Hierarchical Clustering For Fun and Profit

Player comps! We all love them, and why not. It’s fun to hear how Kevin Maitan swings like a young Miguel Cabrera or how Hunter Pence runs like a rotary telephone thrown into a running clothes dryer. They’re fun and helpful, because if there’s a player we’ve never seen before, it gives us some idea of what they’re like.

When it comes to creating comps, there’s more than just the eye test. Chris Mitchell provides Mahalanobis comps for prospects, and Dave recently did something interesting to make a hydra-comp for Tim Raines. We’re going to proceed with my favorite method of unsupervised learning: hierarchical clustering.

Why hierarchical clustering? Well, for one thing, it just looks really cool:

That right there is a dendrogram showing a clustering of all player-seasons since the year 2000. “Leaf” nodes on the left side of the diagram represent the seasons, and the closer together, the more similar they are. To create such a thing you first need to define “features” — essentially the points of comparison we use when comparing players. For this, I’ve just used basic statistics any casual baseball fan knows: AVG, HR, K, BB, and SB. We could use something more advanced, but I don’t see the point — at least this way the results will be somewhat interpretable to anyone. Plus, these stats — while imperfect — give us the gist of a player’s game: how well they get on base, how well they hit for power, how well they control the strike zone, etc.

Now hierarchical clustering sounds complicated — and it is — but once we’ve made a custom leaderboard here at FanGraphs, we can cluster the data and display it in about 10 lines of Python code.

import pandas as pd
from scipy.cluster.hierarchy import linkage, dendrogram
# Read csv
df = pd.read_csv(r'leaders.csv')
# Keep only relevant columns
data_numeric = df[['AVG','HR','SO','BB','SB']]
# Create the linkage array and dendrogram
w2 = linkage(data_numeric,method='ward')
labels = tuple(df.apply(lambda x: '{0} {1}'.format(x[0], x[1]),axis=1))
d = dendrogram(w2,orientation='right',color_threshold = 300)

Let’s use this to create some player comps, shall we? First let’s dive in and see which player-seasons are most similar to Mike Trout’s 2016:

2016 Mike Trout Comps
Season Name AVG HR SO BB SB
2001 Bobby Abreu .289 31 137 106 36
2003 Bobby Abreu .300 20 126 109 22
2004 Bobby Abreu .301 30 116 127 40
2005 Bobby Abreu .286 24 134 117 31
2006 Bobby Abreu .297 15 138 124 30
2013 Shin-Soo Choo .285 21 133 112 20
2013 Mike Trout .323 27 136 110 33
2016 Mike Trout .315 29 137 116 30

Remember Bobby Abreu? He’s on the Hall of Fame ballot next year, and I’m not even sure he’ll get 5% of the vote. But man, take defense out of the equation, and he was Mike Trout before Mike Trout. The numbers are stunningly similar and a sharp reminder of just how unappreciated a career he had. Also Shin-Soo Choo is here.

So Abreu is on the short list of most underrated players this century, but for my money there is someone even more underrated, and it certainly pops out from this clustering. Take a look at the dendrogram above — do you see that thin gold-colored cluster? In there are some of the greatest offensive performances of the past 20 years. Barry Bonds’s peak is in there, along with Albert Pujols’s best seasons, and some Todd Helton seasons. But let’s see if any of these names jump out at you:

First of all, holy hell, Barry Bonds. Look at how far separated his 2001, 2002 and 2004 seasons are from anyone else’s, including these other great performances. But I digress — if you’re like me, this is the name that caught your eye:

Brian Giles’s Gold Seasons
Season Name AVG HR SO BB SB
2000 Brian Giles .315 35 69 114 6
2001 Brian Giles .309 37 67 90 13
2002 Brian Giles .298 38 74 135 15
2003 Brian Giles .299 20 58 105 4
2005 Brian Giles .301 15 64 119 13
2006 Brian Giles .263 14 60 104 9
2008 Brian Giles .306 12 52 87 2

Brian Giles had seven seasons that, according to this method at least, are among the very best this century. He had an elite combination of power, batting eye, and a little bit of speed that is very rarely seen. Yet he didn’t receive a single Hall of Fame vote, for various reasons (short career, small markets, crowded ballot, PED whispers, etc.) He’s my vote for most underrated player of the 2000s.

This is just one application of hierarchical clustering. I’m sure you can think of many more, and you can easily do it with the code above. Give it a shot if you’re bored one offseason day and looking for something to write about.

Forecasting League-wide Strikeout and Homer Rates

Two of the more notable league-wide trends in MLB today are rising home run and strikeout rates.  Strikeouts have consistently trended upward over the past 35 or so years.  Home-run rate, meanwhile, has moved up and down a bit more, but has also increased during that span overall.

An accurate long-term forecast of trends such as these could be valuable.  As this Beyond the Box Score article illustrates, ideal roster construction changes in tandem with the league-wide run-scoring environment.  During periods where offense is scarce, power hitters see their value go up.  When offense is plentiful, speedy contact hitters become somewhat more valuable.

In the following paragraphs, I will attempt to project strikeout percentage and home-run rate — measured as plate appearances per home run — for the 2017-2026 seasons.  First I will take a univariate approach (i.e., use only past patterns in the data to predict future values). Then, I will try to improve the model by adding in an external regressor variable.

Strikeout Rate

First, here’s a plot of the raw data.

Strikeouts rose fairly steadily from the early 1920s to the late 1960s, dipped for about 10 years, then started to tick back up again around 1980.  They’ve been on the rise ever since, and at an especially accelerated pace since 2005.

I considered several classes of time-series models to represent this data, including Auto-Regressive Integrated Moving Average (ARIMA), exponential smoothing state-space (ets), and artificial neural network.  I used AICc to narrow down the field of models somewhat.  I then split the data into a training set and a test set, fit each remaining model on the training data, and evaluated its forecast accuracy based on mean absolute error and median absolute prediction error using a rolling forecast origin.

The data had to be differenced once to make it approximately stationary, after which there was little to no auto-correlation remaining.  Given this fact, it shouldn’t be too surprising that the best-performing model was a random walk with drift.  Below are forecasts from this model for the next decade, along with 80% and 95% prediction intervals.

Year Forecast Low 80 High 80 Low 95 High 95
2017 21.21 20.49 21.92 20.11 22.3
2018 21.31 20.3 22.33 19.76 22.87
2019 21.42 20.17 22.67 19.51 23.33
2020 21.52 20.07 22.97 19.31 23.74
2021 21.63 20 23.26 19.14 24.12
2022 21.74 19.94 23.53 18.99 24.48
2023 21.84 19.89 23.79 18.86 24.82
2024 21.95 19.86 24.04 18.75 25.15
2025 22.05 19.83 24.28 18.65 25.46
2026 22.16 19.8 24.52 18.55 25.77

The model projects a continued, but decelerated rise in K% relative to what we’ve seen the past decade.

Home Run Rate

I used the same general process to fit a model for the home run data, except I first utilized a Box-Cox transformation to stabilize variance.  This time, there was some auto-correlation that remained after differencing.  The best-performing model turned out to be an ARIMA(0,1,1).

Once again, 80% and 95% prediction intervals are given from that model along with the point forecasts.

Year Forecast Low 80 High 80 Low 95 High 95
2017 34.86 31.87 38.58 30.52 40.95
2018 34.86 31.39 39.37 29.85 42.36
2019 34.86 30.98 40.08 29.30 43.66
2020 34.86 30.63 40.74 28.83 44.91
2021 34.86 30.31 41.36 28.42 46.13
2022 34.86 30.03 41.96 28.04 47.32
2023 34.86 29.77 42.54 27.70 48.50
2024 34.86 29.53 43.10 27.39 49.69
2025 34.86 29.31 43.65 27.11 50.87
2026 34.86 29.10 44.19 26.84 52.06

The projection is flat, but with a decrease in home-run rate from one every 32.90 PA in 2016 to one every 34.86 PA going forward.  If plate appearances remain constant, this would mean a 315 home-run reduction across MLB, or just over 30 per team.

Modeling with Regressors

The difficult part with including regressors in the model is finding ones that are known into the future.  Exit velocity, for example, is something that would probably be quite helpful if you were trying to predict home-run rate.  However, since we don’t actually know what it will be in a given season until after that season is over, it doesn’t do much good for forecasting purposes.

One variable I was able to consider was the percentage of home runs and strikeouts in previous years that came from particularly young or old players.  My theory was that if an unusually high percentage of home runs (or strikeouts) came from players that were nearing the ends of their career, league-wide numbers would be more likely to drop in the coming years (and vice versa if  the sources of strikeouts or power were unusually concentrated among young players).

As it turns out, considering age was not especially useful when I back-tested the strikeout model.  Considering the number of old power hitters was not very useful either.  However, percentage of home runs that came from players under 25 was a significant predictor of home-run rate in future years.

I created a variable called “Youth Index” that averaged percentage of home runs from young players in the previous five seasons, weighted by their correlations to home-run rate in the season in question.  To avoid having to forecast Youth index separately, I actually used a slightly different model for each step in the forecast, each considering only known data.  For example, for the 2017 forecast, data from each of the 2012-2016 seasons is available, but for the 2018 forecast, 2017 data is not.  Thus, the Youth index predictor for 2018 used only data from 2-5 seasons back, the 2019 Youth index predictor used only data from 3-5 seasons back, etc.  I limited the forecast to only five seasons ahead, by which point the model started to converge with the univariate forecast anyway.

Year Forecast Low 80 High 80 Low 95 High 95
2017 36.27 33.15 40.16 31.74 42.65
2018 36.25 32.84 40.61 31.32 43.45
2019 36.03 32.38 40.81 30.77 44.00
2020 35.59 31.71 40.77 30.02 44.31
2021 35.67 31.37 41.62 29.54 45.84

*Note: the red and green lines are 80% and 95% prediction intervals just like on the other graphs.  It only looks different because I created this graph manually rather than using an R-package.

The updated forecast projects a more aggressive rebound in PA/HR (i.e., decrease in home-run rate).  The difference overall in the two forecasts is not huge, but not nothing either.  Interestingly enough, the model is over 90% confident that PA/HR will rise to some degree or another next season.

Ultimately, both home run and strikeout rate are influenced by a wide array of factors, many of which are difficult or even impossible to consider in a long-ish term forecast like this.  The confidence bars aren’t quite as narrow as I’d like, which suggests the observed data may end up deviating quite a bit from these projections.  Nonetheless, I think this is a good starting point.

Searching For Overvalued Pitchers

A little while ago, I created a post here about finding undervalued pitchers by looking at improvements between the first and second halves of the season. I had created a linear regression model for the predictions using data from 2002 to 2015, but when trying to use the same model to find overvalued pitchers, it didn’t exactly work as expected (I use the word “work” loosely here — in all likelihood, my predictions will fail as badly as the new Fantastic Four movie). It did find pitchers who suffered massive setbacks, but the majority of those were primarily due to increased — and probably unsustainable — home-run rates.

For example, Matt Andriese had an extremely successful first half of 2016. He put up a 2.77 ERA in 65 innings, backed up by a 2.85 FIP. But those numbers were much like my ex-girlfriend: pretty on the surface, but uglier once you get to what’s underneath. He struck out a lower percentage of batters than the average pitcher during that time while giving up more hard contact. The biggest sign, though, was his deflated home-run rate. He allowed just 0.28 home runs per nine innings, with only 3.2 percent of his fly balls going over the fence. This righted itself in the second half, where his HR/9 increased to 2.15 and his HR/FB to 17.4 percent. On the other hand, he improved his strikeout and walk rates, actually leading to a drop in his xFIP from 4.04 to 3.92 from the first half of the season to the second.

So then what should we expect from Andriese in 2017? The model I created predicts a 5.56 ERA from Andriese, leaning toward his 6.03 ERA from the second half of last season. While it’s unlikely he will allow fewer than 0.3 home runs per nine innings next year, it’s equally as unlikely that he’ll allow over 2 — after all, no qualified pitcher did so over the course of the 2016 season. Andriese’s full-season FIP of 3.78 actually closely aligned with his xFIP of 3.98, so it’s fair to guess that his home-run rates will level out and his ERA in the coming year will be in that range. That would signify an improvement from his 2016 season, rather than his decline predicted from the model.

So, instead of using the model, I took a simpler approach. Here are the players with at least 50 IP in each half of the 2016 season whose xFIP increased the most from the first half to the second:

xFIP Splits
Name First Half xFIP Second Half xFIP Increase
Tanner Roark 3.64 4.83 1.19
Drew Smyly 4.07 5.10 1.03
Hector Santiago 5.05 5.94 .89
Aaron Sanchez 3.41 4.29 .88
James Shields 4.82 5.70 .88
David Price 3.12 3.98 .86

For the purposes of this article, I’ll ignore Santiago and Shields since it’s unlikely that either of them will be relevant in 2017. That leaves four other pitchers whose skills declined dramatically over the course of the season and who you might want to avoid in your drafts.

Tanner Roark

Believe it or not, Roark’s already 30 years old. He’s actually had pretty decent success in his four years in the majors, with a 3.01 career ERA in over 573 innings. On the flip side, over that same time he has a 3.73 FIP, 3.96 xFIP and 4.06 SIERA. That’s not to say he’s a bad pitcher — just perhaps not as good as his ERA would have you believe. The same can’t be said for his second half of 2016. Despite actually bringing his ERA down from 3.01 to 2.60, his already-inflated FIP and xFIP numbers got even worse. His strikeout rate declined by 2.5 percent while his walk rate rose by about the same amount, leading to just a dismal 1.87 K/BB in the second half. His HR/9 nearly doubled as well, but not due to a substantial increase in his HR/FB rate — rather, his fly-ball rate rose from 26 to 37.6 percent, more in line with his pre-2016 career average of 33.9 percent. Why, then, was he able to continue to be successful? A .230 BABIP and a 86 percent strand rate offer an answer. Don’t expect another sub-3 ERA season from Roark — instead, look more toward his Steamer projection of 4.15.

Drew Smyly

For many last year, Smyly was a popular target. He was a high-strikeout guy who was able to limit walks and generate infield flies, prompting Mike Petriello to write this ringing endorsement for him. In his 114 1/3 innings for Tampa Bay before 2016, Smyly had maintained a 2.52 ERA and was among the best at generating strikeouts. But it all went wrong last year. As Tristan Cockcroft points out, Smyly’s season was marked by a first half of bad luck and a second half of deteriorated skills but better luck. His first-half 5.47 ERA was likely undeserved, as he continued getting strikeouts and limiting walks, but was plagued by a .313 BABIP, 63.2 percent strand rate and a 15.0 HR/FB rate, which corresponded to a 4.45 FIP and 4.07 xFIP. His ERA dropped to 4.08 in the second half, but nearly all of his peripheral stats worsened. A move to Seattle won’t fix all his problems, as Safeco Field was actually more hitter-friendly than Tropicana Field in 2016. The sky is the limit for Smyly, but there’s reason to be cautious. It’s possible he bounces back, but this could be who he is now.

Aaron Sanchez

This guy is good, don’t get me wrong. It took a while for some people to catch on, but I was always on his bandwag…all right, so I was one of the guys who didn’t buy in right away. That’s why I don’t do this for a living. Anyway, seeing his name on this list surprised me. After some digging though, it turns out that in my ignorance, I may have been onto something. In 2015, in Sanchez’s trial run as a starter, he was all right. A 3.55 ERA hid a 5.21 FIP and 4.64 xFIP before he got injured and was subsequently moved to the bullpen. When he returned on July 25, he was a completely different pitcher. This time, while he may not actually have deserved his 2.39 ERA, a 3.10 FIP and 3.33 xFIP showed he had made some kind of improvement. Or had he? After all, he only threw 26 innings in the second half of last season. And while there was undoubtedly a huge improvement for him in strikeout and walk rates, something else caught my attention. Take a look at Sanchez’s batted-ball type percentages from 2015:

Pretty clearly, Sanchez improved his batted-ball profile after becoming a reliever. His 2015 second-half ground-ball percentage of 67.6 percent would be the greatest of all of the 1281 qualified pitcher-seasons since 2002, when the statistic started being tracked. His fly-ball percentage of 18.3 percent, while not as extreme, would still rank as the ninth-lowest since 2002. That begs the question: would he be able to sustain those rates when he moved back to the rotation? The answer, as it always is with historically extreme rates, was no:

Both of his rates came crashing back to historically-accurate norms pretty much right away, and they continued to trend in the wrong direction as the season progressed. This, consequently, caused Sanchez’s xFIP to skyrocket. His strikeout and walk rates got worse from the first half of the 2016 season to the second, but only slightly. What really moved his xFIP was his fly-ball rate, which soared (pun intended — maybe I should do this for a living) from 21 percent to 31.8 percent. It’s difficult to say where Sanchez will go from here — after all, this was his first full season as a starter. If he can keep his fly-ball rate at last year’s 25.1 percent — which ranked fourth-lowest among qualified starters — he could still be a pretty decent starting pitcher, even with regression to a league-average HR/FB rate. What’d be even more impressive, though, is if he could keep his batted-ball rates at his numbers from the first half of 2016, which were among the league’s best. Perhaps with a full season under his belt, Sanchez may now have the stamina and endurance to achieve this feat. If he does, look out. If he doesn’t, you’re looking at an average guy.

David Price

Now that I’ve written nearly an entire article’s worth about one guy, let’s talk about another player from the AL East. Price, for much of his career, has been among the elite at the position. Before last season, the only time he had had an ERA above 3.50 was his first season as a starter back in 2009. Every year of his career, he’s been an above-average strikeout guy, but he topped even his own lofty standards when he struck out 27.1 percent of the batters he faced in the first half of 2016. He was unable to sustain that rate, and in the second half of the season he managed to strike out just 20.3 percent of batters, which would have been his lowest full-season rate since 2009. So what changed? Actually, it might have been the first half that was the fluke. Price allowed a 74.2 percent contact rate in the first half, contrasted with a 79.1 percent rate in the second. Those numbers don’t necessarily mean much on their own, but the difference is easy to spot when looking at his career rates:

Price’s whiff rate was higher than ever in the first half of 2016, but it’s tough to figure out why. Per Brooks Baseball, Price was generating swings and misses on his changeup at a career-best rate in the first half, but I couldn’t find any obvious changes to his velocity or movement on the pitch or any other. It’s fair to wonder, then, if his second-half numbers are what we should expect from Price at this point in his career, since his contact rates during that time were much more sustainable. He probably won’t be as bad as his 2016 3.99 ERA, but I wouldn’t be shocked to see it end up above 3.50 for the second year in a row.

Of course, this is not a comprehensive way to find overvalued pitchers. It’s a crude approach, but one that’s meant to highlight guys who fell off in the second half, as they’re the ones more likely to carry over those declined skills into 2017. That being said, xFIP obviously isn’t perfect, and these players all showed that they were capable of posting above-average results over half a season. Take a risk on them if you want, but be warned that they may not be worth the price.

Playing Roulette with Danny Duffy and Wil Myers

January is notoriously slow for baseball activity, but the other week gave us two interesting extensions to digest. Wil Myers was extended for six years and $80 million, while Danny Duffy received five years and $65 million. Both of these players have had interesting careers thus far. Wil Myers has been polarizing in various ways since he was traded for James Shields. The most recent development has been his transition from playing OF to 1B, and seeing if he would be a valuable asset. As for Duffy, he spent part of the season in the Royals’ bullpen before sinker/slidering his way to potential ace status. If you look at both of their production over the last four years you see the following:

Year Duffy Myers
2016 2.8 3.8
2015 1.2 0.6
2014 1.9 -0.1
2013 0.5 2.3
Total 6.4 6.6
Average 1.6 1.7

The table above doesn’t exactly inspire much confidence paying these two individuals the approximate GDP of Qatar. Obviously, the Royals and Padres liked what they saw this past year and were ready to buy into the future. Both Duffy and Myers have youth on their side at 28 and 26 so the teams are buying recent improvements and prime years. Steamer, too, is optimistic about both players, projecting Duffy for 3.1 WAR and Myers for 2.4 WAR.

These deals are not without risk and there is real concern about the inconsistency of both players. As illustrated above, both Duffy and Myers have had years of above-average production and also years where they barely scratched replacement level. These deals may be seen as opportunistic for both player and team, but let’s take a look to see where the value may lie. First, we need to look at how much the team paid and the expected breakeven value.

 Name Contract Value Expected War War Per Year
Duffy 65,000,000 8.1 1.6
Myers 83,000,000 10.4 1.7
Assumes 8M / 1 WAR

Based on this analysis, the teams are paying these players to be exactly what they have been over the past four years. At first glance, this seems like a steep price for the pair who have had middling results but players who have shown superstar upside, even inconsistently, have immense value. A similarity both players share is signing these contracts under team control. Each presumably would have done better on the open market but decided to sell after career years.

Given the significant swings in performance, these contracts are unique because the total value of the contracts may be recouped over 1-2 years. Just this past season, Duffy and Myers were worth 2.8 and 3.8 WAR, respectively. Using the 99th percentile outcome for both these players, a 5 WAR outcome seems to be the absolute ceiling for these two players. Using the same 8M per WAR valuation, a 5 WAR season would produce a value of $40M. This would account for 62% of Duffy’s breakeven WAR and 48% of Myers’. If they were to return to their previous form and be worth 1.6 WAR each year for the remainder of the contract, the team would still enjoy a significant amount of surplus value. If you think 5 WAR is optimistic and prefer to think of their ceiling as closer to 4 WAR, the math still favors the teams’ side of these deals.

Danny Duffy and Wil Myers represent players who offer youth and inconsistency, and they have shown glimpses of stardom. Their respective contracts build both optimism and risk into the final dollar value. The unique part of these deals is quantifying the risk associated with these players. Given their inconsistencies, the teams should potentially expect to receive most of the value in one year while receiving middling results in the others. The Padres and Royals are betting on talent and recent improvements. Teams generally extend players with the idea of receiving consistent year-to-year value. Duffy and Myers portray a more boom-or-bust scenario. Generally, we have an idea of how a contract will go after Year 1; given these two players, we won’t know the result of the deal until the very end.

xFantasy, Part III: Can xStats Beat the Projections?

Last month, I introduced the xFantasy system to these venerable electronic pages, in which I attempted to translate Andrew Perpetua’s xStats data for 2016 into fantasy stats. The original idea was just to find a way to do that translation, but I noted back then that the obvious next step was to look at whether xFantasy was predictive. Throughout last season, I frequently found myself looking at players who were performing below their projection, but matching their xStats production, or vice versa, and pondering whether I should trust the xStats or the projections. Could xStats do a better of job of reacting quickly to small sample sizes, and therefore ‘beat’ the projections? Today, I’ll attempt to figure that out. By a few different measures, Steamer reliably shows up at the top of the projection accuracy lists these days, and so in testing out xFantasy, I’m going to pit it against Steamer to see whether we can beat the best there is using xStats.

First, a quick note on the players included in this dataset. The original xFantasy model was trained on 2016 data for all players with >300 PA. For the comparisons made here in ‘Part III’, player seasons are separated into halves, and all players with >50 PA in a half are originally included. Some have been eliminated due to either changing teams, or lack of data somewhere in 2015 or 2016 (for instance, if they missed an entire half due to injury). Some players have inconsistent names, and since I’m a bad person who does things incorrectly, I indexed my data on player names instead of playerID’s. That means everyone’s favorite messed up FanGraphs name, Kike/Enrique/“Kiké” Hernandez, isn’t included, along with a couple others.

To recap from last time, the inputs I use to calculate each of the xFantasy stats are:

R: xAVG, xISO, SPD*, TeamR+RBI, Batting Order
RBI: xAVG, xISO, SPD*, TeamR+RBI, Batting Order
SB: xOBP, xISO, SPD, TeamSB/PA, Batting Order

(*SPD score has been added to R and RBI calculations since the original xFantasy post)

For both years of xStats data, 2015 and 2016, I’ve separated players into first half (1h) and second half (2h) production. I also have pulled old Steamer projections from the depths of my computer from roughly the All-Star break each year (i.e. early July). All data used today is posted up in a Google spreadsheet here. Anyway, that means our three competitors will be…

  1. Prorated 1h production: Take each player’s 1h pace in the five categories and prorate it to their 2h plate appearances.
  2. 1h xStats (xFantasy): Take each player’s xStats production from the 1h and project the same production over their 2h plate appearances.
  3. Steamer: Take each players Steamer projection and adjust based on actual 2h plate appearances.

Option #1 would be our absolute lowest bar, we should hope xStats can do a better job predicting future performance than the raw ‘real’ stats over that same time period. And I’ll go ahead and say that we’re expecting option #3 is probably the highest bar — Steamer is a much more complex system, using several years of player history (where available), adjusting for park factors, and certainly using many more variables. For xFantasy, it’s just Statcast data, and just over a fairly small sample. This same idea was brought up recently by Andrew:

“Both of these methods use a very, very different process to evaluate players.  xStats uses Statcast data and nothing else, it clings to batted-ball velocity and launch angle. ZiPS is quite different, and there are many resources you can look at to learn more about it.  At the end of the day, though, you see very similar results.  Eerily similar, perhaps.”

– Andrew Perpetua, “Using Statcast to Project Trea Turner”

I hope anyone reading this has already seen that post, as Andrew is using xStats in exactly the way I’m considering here — look at a guy with small major-league sample size, with a recent change in skills (more power for Turner), and see what xStats projects for him.

So first, to set the standard, here are our so-called lower and upper bounds for coefficient of determination (R2) values when predicting second-half (2h) stats:

It’s maybe surprising that using first-half stats does a fairly decent job, but that’s largely due to using the known second-half playing time. Steamer is significantly better across the board, though it’s worth noting that AVG is nearly impossible to predict, with Steamer doing a bad job (R2=.143) but 1h stats doing a far worse job (R2=.067). Before we get to xFantasy, I also wanted to test how my slash-line conversion models were working (i.e. the method used to translate xStats into xFantasy). To do so, I took the rate stats predicted by Steamer (AVG, OBP, ISO) and plugged them into the xFantasy equations to arrive at what I’ll call ‘xSteamer’:

And hey, it looks like very little change. That means Steamer’s relationships between the rate stats and HR, R, and RBI are fairly similar to the ones I’ve come up with. Steamer’s models are still (obviously) better for the most part, though xSteamer somehow beats the original Steamer model when it comes to HR! SB is where we see something completely different, where my model is coming up with significantly worse predictions (R2=.494) than the original Steamer (R2=.671). I would guess that means that historical SB stats are more useful predictors of SB than a player’s current SPD score (actually, a simple check will tell you that’s true, 1h SPD and 2h SPD do not correlate well). In any case, it’s finally time to see where xFantasy falls on this spectrum we’ve set up:

If I’m being honest, I was really hoping to see xFantasy fall closer to Steamer on AVG and HR. But at least for R/RBI, we can definitively say xStats are much more useful for projecting future performance than 1h stats. In the case of SB, it’s a bit of a split decision — xFantasy is doing a poor job, but Steamer does a similarly poor job (both with R2 of approx .49) if using the same inputs as my model.

Now I have to acknowledge an obvious weakness of xFantasy in terms of predictive ability: TeamR+RBI, TeamSB/PA, Batting Order, and SPD…we could likely project each of these much more accurately than just using recent history. Rather than pulling real stats from the first half for each of those, I could have pulled projections or longer historical averages, and likely improved the outcomes significantly. As a shortcut, let’s just eliminate those variables and try again. For this next set of data, I’ve plugged in the *actual* second-half performance for each player in TeamR+RBI, TeamSB/PA, Batting Order, and SPD. For the most direct comparison, I’ll show xFantasy vs. xSteamer:

Now that’s looking pretty good! Gifted with the power to know a few things about actual second-half team performances, xSteamer sets the bar with the highest R2 in each of the five categories. And xFantasy is not far behind! One of the most obvious areas for potential improvement is already a work in progress, with the next version of xStats including park factors. Beyond that, I think this stands as good evidence that xStats could be the basis of a successful projection system, especially if combined with additional historical info or team-level projections. To back that claim up, I’ve come up with one final comparison. Using 2015 xStats, along with the first half of 2016 xStats, we can come up with 1.5 years of xAVG/xOBP/xISO to make predictions of second half 2016. For completeness’ sake, I’ll use a 1.5-year average of all other inputs (i.e. team stats, order, and SPD).

Exciting! It turns out that having more than one half (AKA < 300 PA) of stats leads to much better results. Until we have another year of xStats data to play with, this is the best test we can do for the predictive ability of xStats, but I’m personally quite impressed that this very simple model built on top of xStats is nearly matching the much more complex Steamer system.

At the outset of this whole study, I was hoping to show xFantasy/xStats were at least marginally useful for projecting forward, and I think we’ve seen that. So now I’ll return to the original question: Might xFantasy actually beat Steamer when major-league sample size is small? The easiest possible comparison would be to break down the projection accuracy by player age…

And…yes! xFantasy does a better job projecting the second half for players under 26. Using just Statcast data, 1h SPD score, 1h team stats, and 1h batting order, xFantasy is able to beat the Steamer projection in HR, RBI, and AVG, along with an essential tie in R. The SB model is still quite bad, but I suspect pulling a longer-term average of SPD score (would have to include minors data) would push it up to Steamer’s level. Of course, Steamer is still kicking butt in both the other age ranges. On a mostly unrelated note, both systems do a great job projecting HR/R/RBI for old players, but a surprisingly poor job of projecting SB!

Next time…

So far I’m impressed with how useful xStats and xFantasy can be. I’m looking forward to integrating the further upgrades that Andrew Perpetua has been working on! I’ve also done some initial work on xFantasy for pitchers, using Andrew’s xOBA and xBACON allowed stats, along with Mike Podhorzer’s xK% and xBB% stats. If I can get it to a place of marginal usefulness, I’ll return for a part IV to look at that!

As I said last time, it’s been fun doing this exploration of rudimentary projections using xFantasy and xStats. Hopefully others find it interesting; hit me up in the comments and let me know anything you might have noticed, or if you have any suggestions.

Gary Sanchez Should Bat Second

What do Mike Trout, Josh Donaldson, Dustin Pedroia, Corey Seager and Manny Machado all have in common? Besides the numerous accolades that they share between the Rookies of the Year, the Silver Sluggers, the MVP awards and the combined 16 All-Star appearances, they all share one less obvious trait: they have more career plate appearances batting second in the lineup than anywhere else. Gone are the days of your team’s best player batting third or fourth. The new normal is now MVP-caliber players batting second. It has worked for Pedroia and the Boston Red Sox, Machado and the Baltimore Orioles, Donaldson and the Toronto Blue Jays and Seager and the Los Angeles Dodgers. Not for nothing, but those teams all made the postseason last year with large contributions from their second-hole hitters AND Trout was the AL MVP for the second time in his career on a last-place Los Angeles Angels team. And as more teams continue to adopt this trend, the New York Yankees should also look to bump up their best hitter.

In an appearance the other week on a YES Network interview, GM Brian Cashman has stated that the Yankees have kicked the tires on splitting Brett Gardner and Jacoby Ellsbury in the lineup. This makes a lot of sense when looking at their game; they both rely on their ability to get on base and set the table more so than their ability to drive in runs. Additionally, both players have slowly, but noticeably, been in decline in recent seasons, primarily due to age and injury. Gardner has been the subject of trade rumors over the past few seasons and Ellsbury has been the ire of the New York media for largely failing to live up to the seven-year, $153-million deal he signed before the 2014 season. River Ave Blues has already had a look at how the Yankees would approach this situation and they have provided a solid solution, but they almost immediately toss out the idea of Gary Sanchez batting there for one reason or another, while Sanchez is most deserving of the promotion.

Sanchez has established himself as the Yankees’ most dominant hitter after bursting on the scene last year. The Yankees, their fans, and the nation all expect Sanchez to hit in the third spot in the lineup, a prestigious position considering the history of the franchise, but moving the young slugger to second would not only better suit the team, but would also play to his strengths. Sanchez, despite the short sample size of 231 plate appearances, has proved to be a pretty good fastball hitter. Of the 294 fastballs he has seen, he has connected for a .328 AVG and .781 SLG, and nine of his 20 home runs. Why does this matter? Traditionally, number-two hitters have seen more fastballs than elsewhere in the lineup, and to further cement his commitment to the fastball, per Brooks Baseball, Sanchez had an exit velocity of 94.3 MPH against the heater (Sanchez ranked in the top 10 in overall exit velocity last year). Young players are also traditionally late to adapt to major-league breaking pitches. Can you blame them when they’re up against this or this?

Secondly, it has been proven that two-hole hitters collect more plate appearances per season than the three through nine spots. This is not new information, but the exact number of plate appearances has been up for debate for years. Beyond the Box Score might’ve ended the debate while also examining how the two hole has changed, stating that “[e]ach drop in the batting order position decreases plate appearances by around 15-20 a year,” which might explain why MVPs Trout and Donaldson have made a living there over the past few seasons. An extra 10-20 plate appearances could mean an extra home run or two over the course of the season. Baseball is a game of inches, but it’s also a game of runs.

With a lineup bereft of veteran power and more intent on utilizing the “Baby Bombers,” as they’ve been so aptly named, moving Sanchez up to second could and should give the lineup a much-needed boost if the reliance on Greg Bird and Aaron Judge should go somehow awry. Veterans Matt Holliday, Chase Headley and Starlin Castro have had good seasons and impressive resumes, but they need to return to All-Star form to carry a team of youngsters and a questionable starting rotation. No one really expects Sanchez to produce at the same rate that he did last year, but perhaps a bump up would allow him to produce at an above-average level again.

wRC+ by Leverage: the Good, the Bad, and the Funky

So I got a little carried away with the new splits leaderboard when I was looking up some wRC+ data. I was curious about which players performed the best/worst in high-leverage situations and one thing led to another and it led me to looking at top performers across the three leverage situations (low, medium, and high). If you want to know more about how leverage is calculated there is an old article in The Hardball Times here.

I used the splits leaderboards to gather 2016 hitter data by leverage situation and I only included players who had a minimum of 20 PA per split. Once I gathered all the data I converted each player’s wRC+ by leverage situation to a percentile and calculated each player’s mean percentile rank along with the variation around the mean using standard deviation to produce the following plot.

The blue line is just a LOESS line showing the general trend of the data. What the line is telling us is that players on the extreme end of the percentile ranks also seem to have the lowest variation or, more simply put, good players seem to be consistently good and bad players seem to perform poorly across all leverage situations. Using that plot as my baseline, I started exploring the data to answer some question about player performances in 2016. I included the top 10 players in ordered tables going from from least interesting to most interesting, at least in my opinion. First, let’s look at the top performers from this year.

Players who ranked highest in wRC+ across all leverage situations
Leverage Rank
Name Low Medium High Mean Rank SD
Mike Trout 98 97 99 98 1
Freddie Freeman 97 88 94 93 4.6
Josh Donaldson 97 94 88 93 4.6
Anthony Rizzo 93 88 97 92.7 4.5
Joey Votto 96 98 84 92.7 7.6
David Ortiz 96 99 77 90.7 11.9
Matt Carpenter 91 82 94 89 6.2
Paul Goldschmidt 88 86 88 87.3 1.2
Tyler Naquin 93 81 87 87 6
Ryan Schimpf 80 86 93 86.3 6.5

Boring, Mike Trout leads the way as the top performer. Apparently it doesn’t matter when he comes up to the plate; he is going to smash the ball. But I’m not going to focus on Trout, as I’m not qualified to write about him and he’s above my pay grade, so let’s leave him to the professionals. Like I said before, least interesting first and hopefully it’ll get more exciting as we go. Here’s a fun fact to keep you going: In high-leverage situations among players with a minimum of 20 PA, Ryan Howard led the league in ISO with a 0.640 mark. Ryan Schimpf was second with an ISO of 0.542. And Howard did that with a 0.118 BABIP, too.

Second, let’s take a look at the worst performers of the season.

Players who rated as the worst performers across all leverage situations
Leverage Rank
Name Low Medium High Mean Rank SD
Yan Gomes 17 23 0 13.3 11.9
A.J. Pierzynski 17 21 9 15.7 6.1
J.B. Shuck 25 16 11 17.3 7.1
Nick Ahmed 15 35 3 17.7 16.2
Jake Marisnick 37 19 6 20.7 15.6
Ramon Flores 21 20 21 20.7 0.6
Gerardo Parra 33 29 1 21 17.4
Juan Uribe 19 38 11 22.7 13.9
Adeiny Hechavarria 20 34 15 23 9.8
Alex Rodriguez 19 30 22 23.7 5.7

After a pretty impressive career, although it also came with its fair share controversy, we see A-Rod make this list. And it doesn’t look like he is going to be playing again this year, which casts some doubt on whether he is going to make it to 700 career home runs (he’s currently at 696).  But more importantly, our poorest performer of 2016 looks to be Yan Gomes. I was inclined to say A.J. Pierzynski should actually be considered the poorest performer of the year since his standard deviation was about half of Gomes’, but then I noticed that Yan Gomes was in the 0th percentile in high-leverage situations — literally the worst. Not all-time worst, but still pretty bad! And I guess if you want to argue that the worst percentile should actually be 1, as in the 1st percentile, then you could make that argument, but the value was rounded to 0 when Yan Gomes registered a whopping -72 wRC+ in high-leverage situations. The second-worst was Gerardo Parra at a -59 wRC+; that’s a pretty significant gap between first and second. Fun-fact time: In high-leverage situations, Mike Zunino ran a 30.8% walk rate, although he also struck out 30.8% of the time too. Yasmani Grandal had a 30.4% walk rate to go with a much smaller 13% K%.

Everyone always seems to be looking for players who are on the extreme ends of the leaderboards, but let’s give some love to the unsung heroes of the world, the completely average performers! I wasn’t sure if I simply wanted to use mean percentile rank as a measure for averageness, so I decided to go with what I called Deviation in the table. Deviation is calculated by adding the standard deviations (SD) of a players percentile ranks to the Δ50 column. The Δ50 column is calculated as the absolute value of a players mean rank minus 50.

The most average performers of 2016 in wRC+
Leverage Rank
Name Low Medium High Rank SD Δ50 Deviation
Scooter Gennett 55 46 49 50 4.6 0 4.6
Ezequiel Carrera 46 44 51 47 3.6 3 6.6
Leonys Martin 44 54 47 48.3 5.1 1.7 6.8
Matt Duffy 41 49 49 46.3 4.6 3.7 8.3
Avisail Garcia 45 51 42 46 4.6 4 8.6
Howie Kendrick 46 59 52 52.3 6.5 2.3 8.8
Johnny Giavotella 40 44 42 42 2 8 10
Jason Castro 47 49 62 52.7 8.1 2.7 10.8
Jonathan Schoop 62 53 54 56.3 4.9 6.3 11.2
Brandon Phillips 55 48 63 55.3 7.5 5.3 12.8

And Scooter Gennett comes away as the most average performer of the season! He also ran a 0.149 ISO on the season and I think 0.150 is usually considered average. Look how wonderfully average these guys were; we should all take a minute to enjoy the little things in life. I realize this may not be the sexiest table, but it’s still interesting. You might not be getting a whole lot out of these guys over an entire season, but they are going to go up there and do average things whether you like it or not.

Two tables left — hopefully you’re still with me here. Let’s look at consistency. People always say consistency is key. I guess that’s good advice except when you’re on the terrible end on the spectrum.

Table looking at the most consistent performers based on percentile rank
across the 3 leverage situation (low, medium and high)
Leverage Rank
Name Low Medium High Mean Rank SD
Ramon Flores 21 20 21 20.7 0.6
Ivan De Jesus 32 32 33 32.3 0.6
Mike Trout 98 97 99 98 1
Paul Goldschmidt 88 86 88 87.3 1.2
Johnny Giavotella 40 44 42 42 2
Yunel Escobar 66 69 65 66.7 2.1
Hunter Pence 79 76 80 78.3 2.1
Wilson Ramos 81 80 85 82 2.6
Alexei Ramirez 26 32 28 28.7 3.1
Austin Jackson 43 38 37 39.3 3.2

Ramon Flores and Ivan De Jesus both had extremely consistent seasons; it’s just too bad they are on the wrong end of the spectrum. But I have to say Ramon Flores beats out Ivan De Jesus as he registered on average 12 percentile ranks poorer. In third we see Mike Trout showing incredible consistency while being the top performer in the league, followed closely by Paul Goldschmidt. It’s interesting see the top four players on this list from opposite ends of the spectrum, but the rest of this list bounces back and forth as well.

And here we are, the last one or as the title says “the Funky”. I found that volatility was the most interesting question, or which players showed the most boom or bust in 2016. Most of the players in this list performed best in low- and medium-leverage situations, often above the 90th percentiles.

Looking at players who showed the highest volatility based on percentile rank
across the 3 leverage situation (low, medium and high)
Leverage Rank
Name Low Medium High Mean Rank SD
Sandy Leon 96 84 2 60.7 51.2
David Peralta 95 29 1 41.7 48.3
Dansby Swanson 23 99 15 45.7 46.4
Yangervis Solarte 93 73 5 57 46.1
Mac Williamson 59 95 4 52.7 45.8
Alex Avila 36 99 12 49 44.9
Jarrod Saltalamacchia 41 9 97 49 44.5
Pedro Alvarez 86 85 9 60 44.2
Ryan Zimmerman 21 85 1 35.7 43.9
Kris Bryant 98 91 19 69.3 43.7

After perusing though the list, one of the most interesting names that jumps out should be Jarrod Saltalamacchia and his 97th percentile rank in high-leverage situations last year. And here’s another twist, would it surprise you to hear that in 2016 Miguel Cabrera was the least-clutch hitter among all Tigers qualified hitters? Check out the Tigers leaderboard here. But the 2016 volatility award goes to Sandy Leon, who absolutely mashed balls in low-leverage situations, was no slouch in medium-leverage spots, but dropped off the map in high-leverage situations. I have no idea how BABIP relates to wRC+, but with Sandy Leon it looks like his BABIP reflects what was happening in the different situations (0.434, 0.393 and 0.190). There is probably some combinations of descriptive stats that would explain some of the variance, and BABIP may very well be included, but I’m not going to go into that here.

Hope you enjoyed this. If anyone wants a copy of the R code I used to make the graph and tables, leave a comment below and I’ll pass it along. I ended up finding a pretty cool library to create html tables in R so you don’t have to mess around with formatting and manual inputs. As long as you’re willing to put a little work into understanding css you can basically customize the look of your tables.

The Major Impact of Edwin in Cleveland

Recently, the Cleveland Indians signed slugger Edwin Encarnacion in a bold move to get their formerly middle-of-the-pack offense kick-started.  The deal, which pays $65 million to Encarnacion over three years, can’t be considered a good or bad deal yet — that is still to be determined.  If Edwin, who is, like everybody in the world, constantly getting older, performs like he did for the past two years, then the deal will be a steal for the Indians.  Yet if Edwin begins to show his age at the plate, then the deal will hardly be worthwhile.  Most likely, though, he will accumulate 25-35 (on an overly optimistic side) home runs, while batting for a not noteworthy .280.

Looking over the signing, one can easily come to the assumption that Cleveland will be better with Edwin.  Certainly, any level-headed person wouldn’t consider him to be a minus.  However, nobody has really come out and said that Edwin is the difference between a good team and a great team.  Yet from looking through the depths of Cleveland’s roster, one sees something uncontrollably powerful occurring slowly but surely in Cleveland.  Something that has been in development every since the Indians brought Jason Kipnis to the big leagues in 2011.  And now, with the addition of Edwin Encarnacion, they seem to be done.

What the Indians have done through the past five years is that of a front-office masterpiece.  Last year, they came within a game of winning of the World Series, and this year, they are poised to make a run for the trophy again.  As mentioned before, it all started with the arrival of a noncommittal prospect named Jason Kipnis in 2011.  Kipnis had played well, but definitely not worth a mention in any top-prospect lists.  In the majors, he took a few years to blossom, but he’s been on the rise ever since.  He is now a solid second baseman with speed and power, the second-sacker of every team’s envy.

That same year, Francisco Lindor entered the rookie team of the Indians.  Unlike Kipnis, he became a highly-touted prospect, and his first appearance in the major leagues, in 2015, was widely watched.  And ever since that first game, Lindor has not looked back, joining Kipnis in the ranks of the best middle infielders in the league.

This past year, 2016, was when all the front office’s hard work finally blossomed.  At first, the season did not start out very well.  Stalwart right fielder Michael Brantley got injured early on, and the season’s prospects looked slim.  Yet about a third of the way through the season, something amazing happened.

The Indians were not doing badly, but were definitely not excelling in the season.  So, in a radical move, they decided to see how a prospect would fare in the bigs.  So they summoned Tyler Naquin from the farm system and immediately implanted him in center field.  Thankfully, the lanky Naquin performed above and beyond anyone’s expectations.  He finished the season in the contest for Rookie of the Year, despite missing a good chunk of the season.  Meanwhile, a player who had spent a few years in the bigs yet never really got to play was coming into his own just about the time when Naquin came up.  Jose Ramirez had been drafted by Cleveland after the 2010 season and was called up in 2013.  He didn’t get much playing time, and was sent back down to the minors the next year.  He was called up again in 2015, and played poorly.  However, he wasn’t ready to ruin his big-league career.  At around the time Naquin came up, Ramirez became hot.  He started playing like he hadn’t ever in his career.  Somehow, someway, a switch had been flipped inside him.  Somehow, someway, the Cleveland Indians were in business.

Although the Indians had failed to win the World Series, the season had still been a wild success.  They had built a powerful machine, and with Brantley back in right field for the 2017 season, who knew what could happen?  But still, they seemed to be missing something.  Even with the amazing midseason reinforcements and Cleveland’s powerful lineup (Napoli, Santana, Lindor, Kipnis), the Indians were 18th in the majors in runs scored.  They were getting many runners on base, as their .329 OBP (tied for seventh-best in the MLB) testified.  They just needed one more piece, a guy who could get those many baserunners home.  And although Napoli was big and strong and hit majestic homers, he just wasn’t the guy the Indians needed.  So they signed Encarnacion.  With him on the team and Brantley back, possibilities are boundless.  Their lineup (shown below) will be incredibly potent.

1.  Francisco Lindor;  Shortstop

2.  Jason Kipnis; Second Base

3.  Edwin Encarnacion; DH

4.  Michael Brantley; Right Field

5.  Carlos Santana; First Base

6. Jose Ramirez; Left Field

7.   Lonnie Chisenhall; Third Base

8.  Roberto Perez; Catcher

9.  Tyler Naquin; Center Field

Although the order could be debated on, its potency and presumed consistency are undeniable.  There are only 1.5 holes in the lineup (Roberto Perez=1, Chisenhall=.5), and other than that, the rest of the lineup is stocked with really good players. That’s seven really good players in one lineup.  That is something special.  The lineup is also well-rounded.  There are Lindor, Kipnis, Brantley, and Ramirez providing consistency, while Encarnacion and Santana provide the dingers.  Of course, the four who provide consistency can be relied on to produce at least 15 homers a year.  And although the batting order looks very impressive, the pitching rotation is what really makes the Indians special.  The pitching rotation made it to the World Series minus two of their best pitchers — Salazar and Carrasco — and almost won it!

The addition of Encarnacion will, in my opinion, prove to be great.  The Indians will leap from 18th to fifth in offense in the majors, and they will have a very good regular season.  Again, this is just my opinion, but the Indians do look awfully dangerous come the 2017 season.

Which Pitchers Got Burned on the First Pitch in 2016?

It’s always good to get ahead in the count.  The difference in average run expectancy between a first-pitch strike and a first-pitch ball is over .07 runs.  However, trying to get ahead can go horribly wrong for a pitcher.  With 2016 now in the rear-view mirror, I wanted to take a look back at this past season to see which pitchers got hurt more than most on the first pitch.

First, we have to decide on what it means to be “hurt” on the first pitch.  A pitcher could give up a bunch of singles on the first pitch but not get hurt too bad, depending on the situation.  If a starting pitcher is given a five-run lead, he could give up a few first-pitch hits here and there simply from laying fastballs in the zone trying to get some quick outs.  If he gives up a two-out double with no one on up five runs, the win expectancy isn’t going to change much.  Similarly, in a tie game late with a runner in scoring position, a base hit could give the opposing team the lead for good, leading to a huge change in win expectancy.  Since we are interested in context-dependent numbers, we will use a context-dependent statistic.  Win probability added (WPA) will do the trick.  Adding up WPA will tell us the general story of what happened on the first pitch. Mainly, who got burned.

Using 0-0 count data taken from Retrosheet and Baseball Savant, we will rank the top five relievers and top five starters by their WPA to determine who got hurt on the first pitch in 2016.  However, adding the WPA for all pitchers in 2016 is a daunting and unnecessary task as there are over 600 pitcher stats to comb through.  To get a better sense of the worst performers on the first pitch, we will make a simple rate stat of runs allowed on the first pitch divided by the amount of at-bats the ball was put in play on the first pitch. This stat will allow me eliminate a majority of the pitchers for our list.  If a pitcher has a low to average runs to at-bat ratio, they probably didn’t get burned too much on the first pitch. To make our sample size even smaller, we will use the thresholds of 25 ABs for relievers and 50 ABs for starters to qualify for the data set. This gives us 31 qualified relievers and 48 qualified starters.  Additionally, I have included the at-bats throughout the season that I have deemed to have “hurt the most” for a particular pitcher.  These at-bats led to the largest swing in win expectancy last season for each pitcher.  So, for 2016, here are the pitchers that would like more than a few first pitches back:


#5 Tyler Lyons

(Okay, there were relievers with worse WPAs than Lyons, but when you get hit this hard on the first pitch, I have to make you #5. Hey, it’s my list! I can do what I want!)

Lyons actually had a positive WPA on at-bats where the ball was put in play on the first pitch.  However, most of these ABs happened in garbage time which is the major reason why his WPA is close to average.  He lands on this list due to a .480 batting average on the first pitch, including seven extra-base hits.

2016 breakdown

  • ABs: 25
  • Hits: 12 (3 doubles, 4 home runs)
  • Runs Allowed: 8
  • Runs Allowed per AB: .320
  • WPA: 0.075
  • AB that Hurt the Most: May 6th vs Pirates
    • Tasked with keeping the Cardinals close down one run in the 6th, Jung Ho Kang gives the Pirates a commanding 3-0 lead heading into the late innings
    • Cardinals Win Expectancy before AB: 34%
    • Cardinals Win Expectancy after AB: 13.7%
    • Swing in Win Expectancy: 20.3%

#4 Casey Fien

Fien had rough go of things in 2016, finishing with 6.43 FIP and a -0.8 WAR.  Causing most of that damage were the 13 homers he allowed in just 39.1 innings, including five allowed in 25 ABs on the first pitch.  However, teams will continue to take chances on him as he possesses well above-average spin rate (2,504 RPMs) on his fastball, which makes him a cheap potential bounce-back candidate.

2016 breakdown

  • ABs: 25
  • Hits: 15 (3 doubles, 5 home runs)
  • Runs Allowed: 10
  • Runs Allowed per AB: .400
  • WPA: -0.304
  • AB that Hurt the Most: April 4th at Orioles
    • With no score in the bottom of the 5th,  The Orioles get on the board when Adam Jones doubles off the right-center field wall to bring in two runs
    • Orioles Win Expectancy before AB: 72.9%
    • Orioles Win Expectancy after AB: 85.4%
    • Swing in Win Expectancy: 12.5%

#3 Tony Cingrani

There isn’t a whole lot of mystery as to what Cingrani is going to throw toward the plate.  In 2016, he threw his fastball over 89% of the time.  The first pitch of the at-bat was a huge nemesis for Cingrani last season as nearly half of his runs allowed came as soon as the hitter stepped in the batters box.

2016 breakdown

  • ABs: 25
  • Hits: 10 (4 doubles, 1 triple, 1 home run)
  • Runs Allowed: 13
  • Runs Allowed per AB: .520
  • WPA: -0.547
  • AB that Hurt the Most: May 1st at Pirates
    • Down 3-1, Sean Rodriguez gets the Pirates within one with an RBI triple.  The Pirates tie the game two batters later with an RBI single by Matt Joyce.
    • Pirates Win Expectancy before AB: 19.8%
    • Pirates Win Expectancy after AB: 45.3%
    • Swing in Win Expectancy: 25.5%

#2 Justin Wilson

Good news? Wilson posted an average leverage index of 1.48 in 2016, which was a career high. Bad news? Wilson posted a WPA of -0.84, which was a career low.  The first pitch of the at-bat was no exception.

2016 breakdown

  • ABs: 25
  • Hits: 13 (2 doubles, 3 home runs)
  • Runs Allowed: 10
  • Runs Allowed per AB: .400
  • WPA: -1.108
  • AB that Hurt the Most: August 9th at Mariners
    • Three-run homers hurt.  They really hurt when your team is up three runs in the bottom of the 8th with two outs and one swing later, its a tie game.
    • Mariners Win Expectancy before AB: 8.8%
    • Mariners Win Expectancy after AB: 52.4%
    • Swing in Win Expectancy: 43.6%

#1 Tom Wilhelmsen

Like most of the pitchers on this list, 2016 was a year to forget for Wilhelmsen.  He posted a career low 6.38 FIP and was worth a -1.0 WAR.  Its not all bad, though, as Wilhelmsen pitched better upon his return to Seattle in June, but didn’t pitch well enough to warrant a roster spot at this point for  2017 when he was released in November.  He is currently a free agent.

2016 breakdown

  • ABs: 27
  • Hits: 15 (4 doubles, 2 home runs)
  • Runs Allowed: 10
  • Runs Allowed per AB: .370
  • WPA: -1.149
  • AB that Hurt the Most: May 15th vs Blue Jays
    • Tied a 2 in the 6th, Jose Bautista hits a bases clearing, three-run double to put the Blue Jays in a commanding position to win the game.
    • Rangers Win Expectancy before AB: 47.1%
    • Rangers Win Expectancy after AB: 15.2%
    • Swing in Win Expectancy: 31.9%


#5 Dallas Keuchel

After a Cy Young season in 2015 where he was worth nearly six wins, Keuchel regressed a bit in 2016.  Even with his struggles, Dallas was still an above-average pitcher with a 92 FIP-.  Shoulder issues cut his season short in late August but look for Keuchel to come back stronger next year for what projects to be a strong Astros club, hopefully with better results on the first pitch.

2016 breakdown

  • ABs: 81
  • Hits: 27 (6 doubles, 1 triple, 7 home runs)
  • Runs Allowed: 21
  • Runs Allowed per AB: .259
  • WPA: -0.938
  • AB that Hurt the Most: April 21st at Rangers
    • Here’s that devastating three-run homer again.  Given a 1-0 lead in the top of the first, Keuchel gives the lead right back and then some when Ian Desmond hits the first pitch he sees out to give the Rangers a 3-1 advantage.
    • Rangers Win Expectancy before AB: 44.4%
    • Rangers Win Expectancy after AB: 70.1%
    • Swing in Win Expectancy: 25.7%

#4 Felix Hernandez

Sometimes a pitcher can end up on this list simply due to a larger amount of at-bats against him throughout the season.  This could be the case with King Felix as he had 92 at-bats where the batter put the first pitch in play.  Between the starters and relievers, Hernandez had the lowest batting average against him on the first pitch at .326.  Unfortunately, a majority of these hits came at inopportune times for Hernandez, which influenced his WPA more than most.  Ultimately, this is what we are looking for as timely hitting can hurt a pitcher quite a bit.

2016 breakdown

  • ABs: 92
  • Hits: 30 (8 doubles, 4 home runs)
  • Runs Allowed: 23
  • Runs Allowed per AB: .250
  • WPA: -1.087
  • AB that Hurt the Most: July 20th vs. White Sox
    • One out away from getting out of the first with no damage, Todd Frazier gives the White Sox a huge boost early with a first-pitch three-run shot to left.
    • Mariners Win Expectancy before AB: 50.4%
    • Mariners Win Expectancy after AB: 23.7%
    • Swing in Win Expectancy: 26.7%

#3 Sonny Gray

2016 was an injury filled season for Gray as he had two separate stints on the DL before making a final appearance on September 28th after nearly two months on the shelf.  Injury issues to his right trap and forearm clearly played a role in his poor performance last season.  Did those same injuries also play a role in being hit hard on the first pitch?  While I can’t say for sure, it is possible.  Hopefully those injury issues are behind Gray in 2017 as he looks to return to being one of the best starters in the American League once again.

2016 breakdown

  • ABs: 60
  • Hits: 28 (4 doubles, 5 home runs)
  • Runs Allowed: 17
  • Runs Allowed per AB: .267
  • WPA: -1.322
  • AB that Hurt the Most: May 15th at Rays
    • With the score tied at 1, Brandon Guyer gives a huge advantage to the Rays on the first swing with a three-run shot to left.
    • Rays Win Expectancy before AB: 54.4%
    • Rays Win Expectancy after AB: 82%
    • Swing in Win Expectancy: 27.6%

#2 Kyle Gibson

Right shoulder issues early, home runs and a .330 BABIP could sum up Kyle Gibson’s 2016.  As far as the first pitch goes, homers were again an issue as he allowed eight of them throughout the season. I wouldn’t expect the same in 2017.

2016 breakdown

  • ABs: 75
  • Hits: 35 (9 doubles, 8 home runs)
  • Runs Allowed: 23
  • Runs Allowed per AB: .307
  • WPA: -1.962
  • AB that Hurt the Most: August 17th at Braves
    • With Gibson given a 2-0 lead into the 3rd inning, Freddie Freeman altered the game back into Atlanta’s favor with one swing.
    • Braves Win Expectancy before AB: 26.6%
    • Braves Win Expectancy after AB: 51.2%
    • Swing in Win Expectancy: 24.6%

#1 David Price

When David Price signed last winter with the Red Sox for $217 million, he was expected to be the ace of the staff right away.  That didn’t happen as quickly as Red Sox nation would have liked in 2016 as Price posted his highest FIP- since his rookie season in 2009.  This continued into the postseason with another subpar performance against Cleveland in the Divisional Series.  Ten home runs allowed on the first pitch last season surely didn’t help the below-average season — for David Price standards — in 2016.  With Chris Sale now in the fold to take some pressure off, I fully expect Price to return to being the dominant pitcher we have seen in the past in 2017.

2016 breakdown

  • ABs: 109
  • Hits: 43 (6 doubles, 1 triple, 10 home runs)
  • Runs Allowed: 26
  • Runs Allowed per AB: .239
  • WPA: -2.218
  • AB that Hurt the Most: June 8 at Giants
    • What a time for your first career homer!  With the score tied at 1 in the bottom of the 8th, Mac Williamson unties it on the first pitch of the inning giving the Giants a one-run advantage late.
    • Giants Win Expectancy before AB: 59.3%
    • Giants Win Expectancy after AB: 88.5%
    • Swing in Win Expectancy: 29.2%

**Things to keep in mind**

  • I ranked these two lists by WPA. This doesn’t exactly mean that David Price performed more than twice as bad as Dallas Keuchel on the first pitch. Due to its additive nature, Price could have a higher WPA because he had more opportunities to perform worse than Keuchel.  WPA was just a way to rank the pitchers.
  • There is no predictive nature to WPA. If I were to do this list again for 2017, I would expect a completely different list.
  • This was just a fun list to see who got hit hard on the first pitch in 2016. By no means is this an analysis of why a pitcher performed poorly in 2016. While poor performance on the first pitch certainly can aid in an overall poor season, just because a pitcher performs poorly on the first pitch one at-bat doesn’t mean we can expect him to keep performing poorly the next at-bat.
  • It’s not fun getting burned on the first pitch…


The 2017 Phillies Can Change Baseball Forever

The GM of the Philadelphia Phillies has been accumulating the players to potentially pull off the greatest singleseason heist in the history of baseball.

How will they do this, you might ask?

By utilizing the 3-3-3 rotation.

I will explain why recent rotation alterations by the 1993 Athletics and 2012 Colorado Rockies were not successful. Then I will show how the Phillies version of the 3-3-3 will change the baseball world. But first, let me explain the 3-3-3 rotation and its benefits.

The classic 3-3-3 rotation uses three groups of three pitchers each, pitching once every three games.

Game 1 – Innings 1-3 (Pitcher#1) Innings 4-6 (Pitcher #2)  Innings 7-9 (Pitcher #3)

Game 2 -Innings 1-3 (Pitcher #4)Innings 4-6 (Pitcher #5)Innings 7-9 (Pitcher #6)

Game 3  – Innings 1-3 (Pitcher #7) Innings 4-6 (Pitcher #8) Innings 7-9 (Pitcher #9)

Ideally, each pitcher will throw three innings or 30-50 pitches per appearance. By the end of the season each pitcher will pitch about 162 innings over 54 appearances.

This rotation will help pitchers succeed by:

1) Allowing hitters only one plate appearance against each pitcher

2) Eliminating fatigue by keeping pitch counts down

The more opportunities a hitter has against a pitcher, the better success he has. Dave Fleming of Bill James Online provided statistical evidence from 2008 supporting this fact:


1st PA in G 108606 .255 .328 .398 .727

2nd PA in G 44505 .270 .334 .431 .765

3rd PA + in G 34520 .282 .346 .453 .800

Notice how every hitting statistic increases with each at-bat. To make a few comparisons, Eduardo Nunez was an All-Star last year, and his OPS was .758. All-Star Xander Bogaerts had an OPS of .802. So if you leave a pitcher in past the third AB (generally 7th or 8th inning) you’re facing a lineup full of 2016 Xander Bogaertses. Not exactly a winning formula.

A similar pattern was echoed in pitch counts:


Pitch 1-25 87685 .261 .333 .410 .743

Pitch 25-50 39383 .257 .326 .400 .726

Pitch 51-75 31791 .270 .333 .429 .763

Pitch 76-100 24261 .277 .344 .450 .795

The fact that pitches 1-25 were less effective than 25-50 is due to lineup construction. The rest of the numbers clearly show that pitchers are exponentially worse after the 50th pitch.

In this post, I will explain:

1) Why the 3-3-3 rotation did not work for La Russa in 1993

2) Why the Rockies’ alternative rotation wasn’t accepted in 2012

3) The benefits the 3-3-3 rotation will provide the Phillies in 2017 and beyond

Before we begin, there a few concepts we must accept:

1) Baseball is not archaic; it is ever-changing

2) Categorizing pitchers as only “starters”, “relievers” or “closers” is limiting to the pitchers’ value and abilities. We have to look beyond these inadequate labels. I will use these terms in this article, but attempt to focus on these underlying meanings:

a) Starter – Pitcher trained to throw 5+ innings

b) Reliever – Pitcher trained to throw 1-2 innings

c) Closer – Pitcher with experience throwing the last inning

3) There is no one system that produces winners or losers. You must utilize your personnel to the best of their abilities and limitations

Why the 3-3-3 rotation did not work in 1993

1) The Athletics did not have the personnel to execute the strategy

2) The experiment lasted one week

First, the Athletics had one of the worst pitching staffs in the league in 1993. They were in last place when they implemented the 3-3-3 rotation and had lost nine of their last 12 games. Here is a list of their ERAs in ascending order:

Name                      Training      ERA    Synopsis

Bobby Witt                 SP           4.21     97 ERA +

Goose Gossage          RP           4.53    Age-41 season

Todd Van Poppel      SP           5.04     21-year-old rookie

Ron Darling               SP           5.16       79 ERA+

Bob Welch                  SP           5.29     Age-36 season

Mike Mohler          RP / SP     5.60     Started 9 of 42 appearances

Kelly Downs           RP / SP     5.64     Started 12 of his 42 appearances

Shawn Hillegas      RP / SP      6.97    Started 11 of 18 appearances

John Briscoe             RP            8.03    Started 2 games in 139 IP in career

Only Bobby Witt and Goose Gossage had an ERA under 5.04. Witt was by far their best pitcher and his 97 ERA+ shows he was below average.

The second reason it did not work is the experiment only lasted one week. The public and media backlash from the switch to this rotation was so great that La Russa was forced to abandon the experiment after one week. One week! I don’t care what you do in baseball, if it only lasts one week, then you didn’t give it a real chance. Buster Posey hit .118 in his first week in the MLB in 2009, but the Giants wisely kept him around for 2010.

Why the Rockies’ alternative rotation did not work in 2012

1) They did not have the right personnel

First, let’s describe the specifics of the Rockies’ new rotation. It was a four-man rotation of Jeff Francis, Jeremy Guthrie and rookies Drew Pomeranz and Christian Friedrich. In each start, these four pitchers were given a strict 75-pitch limit. Three rotating pitchers called “piggybacks” would then relieve them.

Game 1 – Francis (75 pitches) Piggyback #1 Reliever #1 Closer #1

Game 2 – Guthrie (75) Piggyback #2 Reliever #2 Closer #1

Game 3 – Pomeranz (75) Piggyback #3 Reliever #3 Closer #1

Game 4 – Friedrich (75) Piggyback #1 Reliever #1/2 Closer #1

Similar to the 1993 A’s, the Rockies made their switch out of desperation. When implemented on June 20th, the Rockies were 18 games below .500 and in a 6-15 slump, on pace to lose over 100 games. Here is a look at the top six Rockies pitcher stats by the end of the year, with ERAs in ascending order:

Name                       Training         ERA       ERA+     IP

Jhoulys Chacin           SP               4.43        105         69

Drew Pomeranz         SP               4.93         94         96.2

Alex White               SP/RP           5.51          84          98

Jeff Francis                 SP               5.58          83          113

Christian Freidrich    SP               6.17          75           84.2

Jeremy Guthrie          SP               6.35          73          90.2

Only one of these starters was even an average pitcher. Three of the four rotation mates were at least 27% worse than the average pitcher in 2012. The issue with the 1993 A’s and the 2012 Rockies are they made these moves in the middle of last-place seasons. They were desperate to change what were the worst pitching staffs in the league. No team heading for a last-place finish is going to respond well to a complete overhaul of the staff in the middle of the summer.

The good news for this particular experiment, however, is that the Rockies pitching staff performed much better after the change was made. In the first 21 games that it was implemented, the starting pitchers improved from a league-worst 6.28 ERA to a league-worst 5.22 ERA. That’s more than an entire one-run improvement! Still the league worst (control your laughter), but that’s a major improvement.

I believe that gives us hope that an alternative and better rotation can be found in the correct circumstances. With the right rotation mates and the correct distribution of pitch counts, I believe there is room for improvement. The key is to train and implement the rotation before the season begins. No pitcher is going to be motivated to try a new system if it is implemented in the middle of a terrible season. It has to be the game plan to begin with, and everyone must be on board. Below you will see why the Phillies have the perfect staff for a 3-3-3 rotation. I have used the 3-3-3 rotation as my basis, but implemented some changes inspired by the 2012 Rockies to ensure success.

How the 3-3-3 Rotation will benefit the Phillies

1) Utilizing the perfect personnel

2) Peak value from assets

3) Health (Physical and Mental)


The Phillies have eight middle-of-the-rotation MLB-ready starters who have demonstrated the ability to get MLB hitters out for multiple innings per appearance. The Phillies have five quality relievers who have demonstrated the ability to get MLB hitters out for one inning+ per appearance. Let’s take a look at the 2016 Phillies stats in order of ascending ERAs:

Name             Training    MLB IP 2016    ERA 2016      MLB service

Asher                 SP                27.2                    2.28              0.061 years

Neris                 RP                 80.1                   2.58               1.104 years

Benoit            RP / CP           48                      2.81                Final Year

Neshek          RP / CP            47                     3.06                Final Year

Eickhoff             SP                 197.1                  3.65                1.045 years

Hellickson       SP                 189                     3.71                Final Year

Ramos             RP                 40                       3.83               0.101 years

Buchholz         SP               139.1             Career 3.96          Final Year

Velasquez        SP                131                       4.12                1.086 years

Nola                  SP                 111                      4.78                 1.076 years

Gomez          RP/ CP           68.2             4.85 w/ 37 SV       Final Year

Eflin                   SP               63.1                     5.54                  0.111 years

Thompson        SP               53.2                     5.70                 0.058 years

Asher, Eickhoff and Hellickson were MLB starters with ERAs under 3.71 last year. Buchholz has the ability to be a front-line starter coupled with a career 3.96 ERA. Velasquez and Nola showed great promise despite rather average ERAs in the 4s. Velasquez sported a 10.6 K/9 ratio while Nola’s curveball has the best horizontal movement in the Majors (9.3 inches, beating out Gerrit Cole). The only two pitchers who disappointed were Eflin and Thompson, two young starters getting their first crack at the majors. Let’s count on them performing better next year.

The best reason why this personnel is perfect is because all of the trained starters have generally similar projections. From a projection and performance standpoint, all of these pitchers are middle- to back-of-the-rotation guys with upside. Nola and Velasquez are projected #2/#3 guys while Eflin, Thompson, Asher and Eickhoff are #3 to back-of-the-rotation guys (Though Eickhoff did have an impressive year in 2016). There is no Kershaw or Verlander or Bumgarner or Cueto who are expected to dominate and throw eight innings every start.

By only allowing them up to 50 pitches and one time through the lineup, the numbers listed in the introduction illustrate that the 3-3-3 rotation puts these players in the best possible position to succeed. Since the numbers are now in their favor, pitchers will have a refined focus and confidence. They can make a structured game plan on how they’re going to attack each hitter. This will limit extended innings under duress and ultimately build confidence in the minds of these young pitchers.

You may ask, Kevin, the Phillies aren’t going to contend in 2017. Why go through such a drastic change to get marginally better?

The answer is using the 2017 season as a stage for their assets to increase in value.

Asset Valuation

The Phillies are not in line for a winning season in 2017. They most likely won’t win 80 games in 2018. But 2019 is their year. That amazing 2018-2019 class of Kershaw, Donaldson, Machado, Harper, Pollock, LeMahieu, Keuchel, Harvey, Wainwright, Corbin, Smyly and Shelby Miller will be theirs for the taking, as the only money they have tied up is to Odubel Herrera. Even the 2017-2018 class of Arrieta, Cobb, Darvish, Duffy, Pineda, Tanaka (option), and Cueto (option) could insert an ace or #2 into their staff.

That is why they need to act now. They must increase their pitchers’ values now and acquire better assets with 2019 in mind. The free-agent market will be booming from 2017-2019, thus lowering trade-market value of any player after this year’s deadline. Instead of trading away prospects to get the guys they need, teams will simply open their pocketbooks. Now is the time to trade these middle-of-the-rotation guys away. Especially because they are not all in the 2019 plans.

“Utility Pitchers”

What is the most overpriced asset on the market right now? Relief pitching. More specifically, pitchers who can pitch multiple innings in relief in tough situations. See: Andrew Miller, Kenley Jansen, and Aroldis Chapman. By utilizing the 3-3-3 method, you are training your starters to pitch multiple innings in different scenarios and relieve in later innings. The 3-3-3 method trains your pitchers to achieve the greatest possible value by becoming what I like to call “utility pitchers.”

What makes players like Ben Zobrist, a .266 career hitter, and Ian Desmond, a .267 hitter, worth $60-70 million? They are utility players. Teams these days love utility players and are willing to pay big money for them. They are more valuable now than they have been in all of history. The same can be said for utility pitchers.

If you have ever been to the Arizona Fall League, it is used as a stage for the game’s top prospects. Starting pitchers generally pitch three innings, and relief pitchers will pitch 1-2 innings each for the remainder of the game. They do this to give teams’ top minor-league players exposure to higher competition with an added benefit of raising prospect value in the eyes of other teams. By sending their players to compete with top minor-league competition for all scouts to see, a good showing will raise potential trade interest. For example, this year the Giants sent a young catcher named Aramis Garcia, a former second-round pick. Garcia doesn’t fit into the Giants MLB plans with a player like Buster Posey entrenched at catcher until 2022, but they used him as one of their eight player selections anyway. I can surmise they did this to boost his stock for potential trade scenarios. The Phillies do not have all their current pitchers in their 2018-2019 MLB plans, so why not show them off to other teams?

By using the 3-3-3 method in the MLB as a stage for their abundance of young pitching talent, their pitchers will:

1) Get experience against the top talent in the world

2) Potentially increase their trade value

3) Limit innings to 130 – 160 IP

4) Give young pitching the best chance to succeed at the MLB level

5) Keep their innings down and arms fresh

The Phillies 2017 3-3-3 rotation, which you will notice is a quasi version of the 3-3-3 that I referenced above, would look like this:

1st Group – Hellickson (3) Asher (3) Eflin (2) Neris (1)

2nd group –  Nola (3) Eickhoff (3) Thompson (2) Gomez (1)

3rd Group –  Velasquez (3) Buchholz (3) Benoit (1) Ramos (1) Neshek (1)

Why this particular grouping?

1. Ability to sell three of what we call “closers” at the deadline. They can also switch Benoit and Ramos to the closer role on any particular day, giving Klentak five pitchers with closing experience to sell.

2. Give Eflin and Thompson only 2 IP per appearance because of their struggles last year. This should increase their confidence by decreasing their perceived pressure.

3. Since the Phillies signed two relievers to one-year deals in the offseason, it is apparent that Klentak wants to sell them off at the deadline. This is why I chose the quasi 3-3-3 system.

Imagine Klentak’s bargaining power at the deadline if he has even three of these newly trained utility pitchers pitching well, especially if one is a guy like Asher, Eflin, or Thompson? He could promise 5+ years of control of a utility pitcher who can be a traditional starter or a multi-inning reliever out of the bullpen.

Some people will read this and think that this would be a “demotion” or “devaluation” from being a “starter.” This is not true. All of these pitchers made it to the MLB as what you would call “starters.” They have excelled at pitching 6+ innings per game. This experiment would simply add value to all of them. Just as playing Ben Zobrist at LF, RF and SS doesn’t take away his ability to play 2B.

Most relief pitchers don’t get drafted as closers or relief pitchers. They are given chances at various roles and stick with whichever role suits their strengths best. Look at Chapman and Andrew Miller. Look at Joe Blanton! Terrible pitcher as a labeled “starter” but excelled in a set-up role for the Dodgers last year. General managers won’t trade for a guy for a postseason run if he hasn’t proven that he is going to be a solid contributor in the specific role they need for their team. So by using 2017 as a value-booster, you train all of your pitchers for multiple roles so you can have the leverage to trade any of your guys to any team. Every postseason team needs pitching. The 3-3-3 rotation will give Klentak unlimited options to acquire talent that will help the 2019 team be successful. GMs are most vulnerable at the deadline, and it is time to take full advantage.

Some people might argue that bringing up all of these pitchers at once would be a waste of MLB service time. But what is more important to a GM who has multiple pitchers with middle-of-the rotation ceilings? An option year or service time? This experiment is exactly that, an experiment. It is a trial run for one half of a season to ramp up current asset valuations to acquire a lot of quality pieces for the future. Since all of these pitchers are already on the 40-man roster, sending them to the minors would waste an option year anyway. So why not give this a try? The worst thing you could lose is half a season of MLB service time on a few guys who have served less then 20% of one year in their career.


In an arm-health study by Dr. James R. Andrews the following chart is comprised:

Ages 14 and under – 66+ Pitches (4 days rest) 51-65 (3) 36-50 (2) 21-35 (1) 1-20 (0)

Ages 15 and over – 76+ Pitches (4 days rest) 61-75(3) 46-60 (2) 31-45 (1) 1-30 (0)

These pitchers are prized assets. Millions of dollars coupled with thousands of hours of prep, coaching and playing time are used per arm. Why don’t we take better care of these players?

As a kid, your parents told you to eat your vegetables, sleep eight hours a night and stay in school while getting 60 minutes of exercise a day. But as we grow older we continually skip our vegetables, sleep five or six hours a night, forget to keep our brains active, and rarely exercise. We feel that we can still function this way, but more importantly, we feel we have to function this way. This is because we put too many responsibilities on ourselves at the expense of our own well-being. I’m arguing that we are giving these pitchers too many responsibilities, at a detriment to their peak physical health. Why? Because traditional baseball knowledge tells us that a five-man starting staff is the right way to go in 2017. But look back at history: there used to be one-man, two-man, three-man and even four-man rotations. Those proved to be unsuccessful. I am saying that the five-man rotation isn’t working either. It’s time to make a change.

What if we treated these valuable multi-million-dollar arms with the care that we take with our Little League arms? I propose a hopeful plan of three innings finished for each starter, but an absolute maximum of 36-50 pitches no matter what. These pitchers will then receive two days of rest for every 36-50 pitches, thus receiving the care a child under 14 would receive (see chart above). It is impossible to argue that this wouldn’t be a healthier system than the one we have now. Finally, let’s shift back to trade value. If Klentak is making deals on July 31 and a playoff contender is asking him how his players can help them win a championship, health is another big concern! If he can say that his pitchers have been put on a stricter regimen than any other team in the league, and that his players’ arms are healthier and more fresh than any other team in July in the history of baseball, that is going to increase his bargaining power. Remember, keeping players healthy, putting them in the best position to succeed and increasing trade value all are focused on the 2019 season. Klentak’s initial plan has always been focused on the 2019 season. And this plan will add tremendous benefit to that goal.


Now I am not saying that every team should utilize this strategy. I am not saying this is the future of baseball for eternity. I am saying that with the Phillies assets, at the perfect time in their development, this will be a great strategy to use. A Double-A or Triple-A prospect is worth much less than an MLB-proven prospect. A pitcher who can relieve, start and spot-start is worth more than just a conventional “starter” or “reliever.” More utility is always better than less utility. Healthier arms are better than overused arms.

I am saying the Phillies should give this a try for half of a season in which they won’t win more than 80 games. There is nothing to lose. And hey, if everything goes to plan, maybe this starts a revolution. If not, then they seamlessly revert to a five-man rotation in August. The goal of business is to buy low and sell high, looking for the most reward for the least amount of risk. This is about as high-reward as you can get in a sub-.500 season with about as little risk as I can imagine.

A new idea is always crazy before it makes sense. In the 1920s and 30s it was a rule that star pitchers had to throw 10-20 relief appearances in addition to their normal starting roles. In the 1880s, catching a ball on one bounce was an out. It even used to be legal for a first baseman to grab a runner by the belt so he couldn’t steal second! It is time for a new discussion about the modern-day pitching staff. It is time for rebuilding teams to try new things to get an edge on the competition. It is time for the game of baseball to go through yet another change. We owe it to the fans, to the players, and to the history of our beloved game. We owe it to ourselves to put our reputations on the line for the greater good of baseball.