Author Archive

Cashman Strikes Again

Brian. Cashman.

Probably one of the more unpopular figures in New York sports (for no good reason) has struck again and made a great deal for the Yankees.

I’m a Cashman defender, through and through. I think that 15 playoff appearances in 18 seasons, 6 pennants, and 4 championships is a stellar résumé, and that he doesn’t get enough credit for it, regardless of how big the payroll is, or how much control he really had early on. We’ve seen Cashman at his best these past few seasons. Sure, they haven’t been the most successful for the Yankees; anything less than a championship is a failure. That’s the Steinbrenner way, and the way it should be. But we’ve seen that Cashman can operate without an infinite payroll and can make moves other than opening the Yankee’s checkbook.

You can criticize Cashman for some free agent signings, sure. You can say he didn’t have to do much to win his first championships and inherited a great roster; that’s more than fair. You can’t, however, deny that Cashman is a master of trades, especially recently. He understands trading from a surplus as well as anyone. And he does so to acquire either low-risk, win-now pieces, or young, talented, potential-filled, cost controlled pieces.

And with this latest trade of Adam Warren and Brendan Ryan for Starlin Castro, Cashmoney has done it again!

Before we dive into this latest case of Cashman genius, let’s highlight some of his recent gems.

Catcher JR Murphy to the Twins for outfielder Aaron Hicks, November 2015- I like Murphy a lot. He seemed like a great guy, and was a good player, but the Yankees were loaded with catchers. McCann is locked in for 3 more seasons and is basically immovable with his contract. Gary Sanchez has been destroying the Arizona Fall League and seems like a legit power bat. Austin Romine is somewhat boring, but has the skills to be a serviceable backup catcher, and many are still high on Luis Torrens. He missed last season with an injury, but is still viewed as a solid prospect. Murphy was merely a commodity. Hicks, once a top prospect, is a switch-hitting 26 year old outfielder, and seemed to turn the corner last season, posting the highest average, slugging percentage, wOBA, and WAR of his career. He hits lefties very well too, batting .307 against southpaws last season. At worst, Hicks is an above average fourth outfielder who will play very solid defense, and provide speed and right handed pinch-hitting ability off the bench. At best, he has a similar career turnaround to Carlos Gomez (as Paul Sporer notes), and becomes the All Star outfielder he was once expected to be. He’s under team control through 2019.

Infielder Martin Prado and RHP David Phelps to the Marlins for RHP Nathan Eovaldi and 1B/OF Garrett Jones, December 2014- Garrett Jones was obviously a bust but this was still a great trade for the Yankees. A lot of the Yankees trades have worked out, but a good way to look at trades (or life in general if we really want to get deep here) is to not be so results-oriented. Use your resources to make the best decision possible, and let the pieces fall where they may. The Yankees traded a fringe starter/long reliever and aging utility man for a power lefty off the bench and a promising, young, flame-throwing arm. And it worked out really well. David Phelps is David Phelps, and Martin Prado wouldn’t have won the Yankees the pennant last season. Nathan Eovaldi looked like a changed man under Larry Rothschild’s guidance last season. Adding a nasty split to go with his high-90’s fastball has done wonders for him. He’s able to change hitters’ eye-levels more effectively, meaning more strikeouts and less hard contact. It paid off last season. He won 14 games and got better as the season progressed, striking out 8 batters per 9 innings, and posting a 3.67 ERA in the second half. Eovaldi looks like he’s on his way to being a solid 2-3 starter, and is under team control through 2017.

RHP Shane Greene to the Tigers, received shortstop Didi Gregorius from the Diamondbacks, December 2014– This may go down as one of the better trades the Yankees have ever made. Big statement, I know. But the Yankees traded a barely major league starter for what looks like the shortstop of the future. At the time, Greene was viewed as a back of the rotation starter at best. Fast forward a year and he posted an ERA of nearly 7, and may not have a job next season. Gregorius, meanwhile, overcame a slow start and turned into one of the Yankees most valuable assets. He hit .294 with a .762 OPS in the second half, ranked as the 4th best shortstop in baseball(!) per fWAR, and played stellar defense. Jorge Matteo is on the horizon, but Gregorius is the shortstop of the present and the future if he keeps this up. I envision a breakout season coming for the Dutchman. He’s under team control through 2019.

All 3 of those trades were low-risk deals, dealing from a surplus for high-upside guys under team control for the foreseeable future. Two have worked out very well, and the third has a great chance to.

(Side note: When I was looking through for these exact contracts, and was reminded that Jacoby Ellsbury is under contract through 2021, I almost threw up.)

Cashman, however, has also showed he can make great trades in season for win-now players. In 2014, he gave up Vidal Nuno for Brandon McCarthy. They lost McCarthy that offseason, but Nuno is no Clayton Kershaw and McCarthy was great for them down the stretch, performing at an ace-like level; he had a sub 3.00 ERA. The deal he made for Chase Headley that season was similar. While fans may be understandably upset at it now, as Solarte had a solid season last year, and Headley, for the most part, did not, it was a really good trade at the time and still could be for the future. Solarte had just a few months of MLB experience and while he started off hot, he was drastically slowing down. Headley had a 31 home run season, All Star appearance, and Gold Glove under his belt. The Yankees locked Headley up this past offseason through 2018, and will want him to improve. Solarte actually had a higher average, wOBA, wRC+, and WAR last season. Headley, though, has proven he can be an All Star player, and he should rebound this season, especially defensively. And the point is, at the time, it was a really smart move. The Yankees also included Rafael De Paula in that deal, but has done nothing of note and is still floating around the minors.

I also feel obligated to mention the Kelly Johnson for Stephen Drew trade in 2014 with the Red Sox. Probably the most hilarious trade I’ve ever seen for so, so many reasons, but I digress.

Back to the deal at hand. Warren and Ryan for Castro. A fringe starter/middle reliever and a veteran, light-hitting (to be generous), backup middle infielder for a 25 year old, 3-time All Star, once top prospect, yet still very promising middle infielder. I know that sounds too simple, but that really is what this deal is.

Brendan Ryan… let’s just get this out of the way early. This is addition by subtraction for the Yankees if we’re being honest. He’s a .234 lifetime hitter with 19 home runs in 2,872 plate appearances. I wish I was making that up, but I’m not. And for a guy who supposedly has a great glove, I saw him make/not make a number of questionable plays last season. He was wasting a roster spot. Now his role can go to someone more promising like Dustin Ackley or Rob Refsnyder.

Parting ways with Adam Warren isn’t easy, but it’s not the end of the world. Warren is a good pitcher. I liked watching him grow these past few seasons and wish the Yankees could have kept him in the rotation last season, where he was very reliable, although he did seem to also find a niche as middle/late inning reliever. Warren “knows how to pitch” to use a cliche. He doesn’t blow you away, but he works his fastball in with his offspeed stuff well, locates his pitches, and gets people out. Pitchers usually improve when they move to the NL, and with this opportunity to finally be a starter, I expect Warren to be a nice addition for the Cubs next season. He should be a solid back of the rotation starter, or reliable reliever if they go that route. Not sure Adam Warren is the key to ending their century long World Series drought, but he doesn’t hurt them, that’s for sure.

The idea that the Yankees are giving up some Cy Young caliber starter, however, is absurd. Like I said, Warren is a 4-5 starter and the Yankees just don’t have room for him. Tanaka, Severino, Eovaldi, and Pineda are all rightfully ahead of him. And Sabathia, Nova, and Mitchell are in the mix as well. Plus, it’s almost a certainty that the Yankees will add another starter this offseason, so there really was just no room for him in the rotation. He would’ve been a nice arm in the bullpen, as the Yankees need another right-hander out there so Dellin Betances can rest more, but he can be replaced. There’s always plenty of right-handed relievers available on the free-agent market, and the Yankees have some in-house candidates as well. This does, however, make trading Miller even dumber, but that hopefully shouldn’t be happening anyway. The bottom line is, while Warren is a nice pitcher, he was no more than a middle reliever/depth starter for the Yankees. Turning him into Starlin Castro is gold by Cashman.

Speaking of Castro, let’s get into the real headliner of the trade. Castro will be just 26 at the start of the 2016 season. He’s been an All Star, he’s got a proven bat, and while he’s had some troubles, he seems to be trending in the right direction. He’s got a career slash of .281/.321/404. His wOBA is .316 and his wRC+ is a below-average 96. Every full season he’s played, he’s hit between 10 and 14 home runs. Now, none of these numbers set the world on fire, but there’s been very promising stretches sandwiched in there, and when you consider he’s a middle infielder, these numbers look much better.

The problem with Castro has been inconsistencies. Last season he was not very good offensively, and he wasn’t in 2013 either. But he had very promising seasons the rest of his career; 2010, 2011, 2012, and 2014. Over those 4 years, his wRC+ was 99 or higher each season, he’s hit at least .283 in all of them (over .300 twice), and has averaged a WAR of 2.9 per 162 games. Those are really good numbers, especially for a middle infielder. We can’t completely discount his two poor seasons, but a change of his scenery may be all he needs, as the talent is clearly there.

Along the lines of a change of scenery is his position change from short to second in August of last season. In his 33 games at second base last year, he hit 5 home runs, drove in in 22 runs, and had a slash of .339/.358/.583. Small sample size, yes, and logically there should be no relationship between what position a guy’s playing and his ability to hit, but maybe that’s the case for Castro. If he feels more comfortable at second, it could be enough to get his mind right and allow his talents to take over. He’s still before his prime, and he’s under team control through 2020 for an affordable salary. It’s a risk the Yankees needed to and can take. Stephen Drew just had the worst year and half I’ve ever seen, and while Ackley and Refsnyder have potential, they’re not as good as Castro. Ackley can now have a utility role, while Refsnyder can perhaps be used as part of a package deal to get a frontline starter.

On a larger scale, this trade shows the genius of Cashman’s trading ability, and the Yankees’ continued win-while-rebuild mode. They’re keeping their top prospects, yet still are getting younger and more athletic by trading from areas of surplus and buying low and selling high on players. For all the talk of the Yankees being an old, veteran team, they very quietly are assembling a great, young core. Gregorius, Castro, Greg Bird, Aaron Judge, Matteo, Refsnyder, Hicks, Severino, Eovaldi, Pineda, Tanaka, and Betances are all either top prospects, proven players, potential budding stars or somewhere in between. Not one of them is older than 27 and all are under team control for the foreseeable future. With this young core, and a ton of money coming off the books very soon, the future is bright in the Bronx. You can thank Brian Cashman for that.

Introducing Two New Pitching Metrics: exOUT% and exRP27


In the early 21st century, Oakland Athletics’ General Manager Billy Beane revolutionized baseball forever. He was the first general manager in baseball to heavily utilize sabermetrics in his baseball operations. This isn’t a history lesson though, I bring him up because of his idea that outs are precious, and as a hitter your goal is to not make out, thus him prioritizing OBP so heavily. In the following years, baseball statistics have seen phenomenal progress on both offense and for pitchers. While I believe FIP and xFIP are both very useful statistics in really measuring a pitcher’s skill, my problem is that they essentially ignore all the batted ball data that we have (GB%, FB%, LD%). SIERA and tERA have solved some of these problems, but are far from perfect, and I believe the more statistics we have, the better.

As I mentioned with Beane, while we largely focus on a hitter’s ability to not make out, we still don’t have a catch-all statistic to realize how effective pitchers are at getting batters out, because if the batter’s goal is to not make out, the pitcher’s goal is to get the batter out. So I present to you expected out percentage, or exOUT% (the name is certainly a work in progress). exOUT% sets out to answer a simple question: For any plate appearance, what is the likelihood that the pitcher will get the batter out? This can easily be found by just looking at a pitcher’s opponent OBP, but that is rather primitive, and we can get a better estimate by focusing more on pitchers’ skills to strike people out, not walk batters, and the type of contact they are giving up, and also trying to negate the effect of the defense by him, by just using league averages. So to calculate a pitcher’s exOUT%, I used K%, BB%, GB%, LD%, FB%, lFFB%, and 2014 league averages on ground balls, line drives, and fly outs. (HBPs are essentially ignored but can certainly be incorporated in a future version, this is pretty much exOUT% v1.0)

I want to give full disclosure, I am not a statistician or close to it. Math and statistics are an area of interest and I am currently pursuing a degree in math-economics, but I am far from a professional, so I recognize there are going to be errors in my data. This is an extremely rough version; there’s even a combination of data from this year and last year so there will be inconsistencies, as I don’t have the resources to gather all the data I need. If after reading this, you are interested in this and would like to take this further, please feel free to contact me if you have the skills necessary to advance this further (or even if you don’t).

I will first post a simple step-by-step breakdown of how to calculate exOUT%, and then get into more detail and take you through it with Clayton Kershaw, because well, he is awesome.

1- Add K% and BB%, subtract this percentage from 100%, this leaves you with a balls in play%, let’s just say BIP%

2- Multiply the pitcher’s GB% (make the percentage a number less than 1, for example 40% is .4) and BIP% (leave it between 1 and 100, ex 40%), this gives you a GB% for all PAs, not just balls in play, we’ll call this overall GB%, or oGB%… now multiply this percentage (in between 1 and 100) times the league average percentage of ground balls that don’t go for hits (league average is .239 on ground balls in 2014, so out percentage on ground balls is 76.1%, but make it .761…. this will give you a percentage you can leave between 1 and 100, if the number is 20%, that means that there’s a 20% chance that pitcher will induce a ground ball out that PA, assuming league average defense, we can assume this because we’re using the league average for batting average on groundballs… we’ll call this exgbOUT%

3- Now follow the same steps but with LD%, exldOUT%, the percentage chance for any given PA that the pitcher will produce a line drive out. (The league average on line drives last season was .685 (!) so that means there is a 31.5% chance a line drive will result in an out)

4- Same thing with FB%, sort of, because we also want to incorporate IFFB%. So multiply a pitcher’s FB% by their IFFB%, this gives you the percentage of balls in play that the pitcher produces an infield fly ball (bipIFFB%). Multiply this percentage by their BIP% to get his overall percentage of PAs that result in an infield fly, and this will also be their exiffbOUT%, because any infield fly ball should be converted to an out, and if not, it’s to no fault of the pitcher, so we won’t punish him. Next subtract a pitcher’s IFFB% from 1 or 100, whatever, and this is their balls in play percentage of fly balls that are normal fly balls, to the outfield. Multiply this number by their BIP%, this gives you the overall normal FB% for a pitcher, not just balls in play. Multiply this number by .793 (the league average on fly balls in is .207, so there’s a 79.3% that a fly ball will result in an out). This number is the percentage chance that for any given PA, the pitcher will produce a fly ball out to the outfield. Add this exnfbOUT% (n for normal) and his exiffbOUT% and you have his exfbOUT%, the percentage that for any given PA, the pitcher will produce a flyball out, to the infield or outfield.

5- Add K% + exgbOUT + exldOUT + exfbOUT

6- You have your exOUT%


The terms are not that technical or scientific so I don’t confuse anyone — I tried to simplify a very complicated procedure as much as possible. To clarify and give you an example, let’s go through Clayton Kershaw.

Kershaw profiles like this (I compiled this data on 8/21): 32.3 K%, 4.9 BB%, 52.8 GB%, 26 FB%, 11.8 IFFB%, 21.2 LD%.

So let’s look at the balls that don’t go in play, strikeouts and walks. Add the two and balls not in play percentage is 37.2, 4.9% are walks and thus won’t be an out, and 32.3% are strikeouts so will be an out. Thus far, Kershaw’s exOUT% is 32.3 (of a possible 37.2 so far)

Now let’s look at the balls in play. People will usually say that a pitcher can’t control what happens when a ball is in play, but I vehemently disagree, the type of contact the pitcher gives up can’t be ignored and largely effects what will happen to the ball in play. I will quote a FanGraphs article here to explain it, “Generally speaking, line drives go for hits most often, ground balls go for hits more often than fly balls, and fly balls are more productive than ground balls when they do go for hits (i.e. extra base hits). Additionally, infield fly balls are essentially strikeouts and almost never result in hits or runner advancement.” And FanGraphs also gives us this data from 2014.

GB: AVG- .239, ISO- .020, wOBA- .220

LD: AVG- .685, ISO- 190, wOBA- .684

FB: AVG- .207, ISO- .378, wOBA- .335


So this means that fly ball pitchers are most likely to get outs, although they may be less effective because when they don’t get outs, it’s more trouble than for ground ball pitchers. But remember, this statistic is just finding the chance that the pitcher will get a hitter out.


All right, so, let’s calculate Kershaw’s exgbOUT%, exldOUT%, and exfbOUT%; you can follow the numbers along with the steps I listed above.


GB%- 52.8

62.8 x .528 = 33.1584

(33.1584 x .761)=  25.23354424 exgbOUT


LD%- 21.2

62.8 x .212 = 13.3136

(13.3136 x .315) = 4.193784 exldOUT


FB%- 26

26 x .118= 3.068 bipIFFB%

26 x .882= 22.932 (bipFB%)

62.8 x .22932= 14.401296 (onFB%)

14.401296 x .791= 11.3914251 exnfbOUT%

62.8 x .03068= 1.926704 oIFFB% and exiffbOUT%

exnfbOUT% + exiffbOUT% = 13.3469317 exfbOUT%, if you followed my math exactly a decimal may be off, like 13.31 something, but this is the number the excel doc chugged out, so I’m trusting that, my iPhone calculator can’t carry all the decimals sometimes.

Now add them all up

32.3 + 25.23354424 + 4.193784 + 1.926704  + 11.3914251 = 75.07%

K% + exgbOUT% +  exldOUT% + exiffbOUT% + exnfbOUT% = exOUT%

The league average exOUT%, using league average statistics from 2014 for the ones involved, is 69.8%. Scherzer leads the majors (well the 89 pitchers I was able to export data from FanGraphs) with a 76.43 exOUT%. If you want to look at it as a more concise and better version of opponent OBP, his is .236, so, you know, good. Here is a picture of the data for the top 37 — the J column is what you are looking at. Betances is in their because I wanted to calculate one reliever. 

View post on

All right, I’ve explained it a bit in the prologue, but now that you’ve seen it, let me explain more why I like this stat. Well first, I created it and calculated, so, well, yeah… but I also like this stat because it answers a very simple question “How good is a pitcher at getting people out?” Pitching in its simplest form, is exactly that, getting people out. The stat recognizes that there’s basically only these outcomes for an at bat: strikeout, walk, ground ball, line drive, and fly out, and looks at the pitcher’s stats in these categories to determine how many people he should be getting out. The stat is more predictive than evaluative in nature, because you can calculate a pitcher’s actual out percentage, but that doesn’t nearly tell the whole story, because a lot of luck is involved with balls in play, and other fluky outcomes.

This operates under the basis that a ground ball will perform the way the average ground ball does, a line drive performs the way an average line drive does, and a fly ball behaves the way a typical fly ball does. There could be guys getting very fortunate with ground balls: having a great infield behind them, balls not squeaking through the holes; with line drives: being hit right at people; and fly balls: staying in the park, having outfielders who cover a lot of ground. And there could be guys who are getting unlucky: the ground balls are getting through the holes, the infielders don’t have range; line drives seem like they are always going for hits, and fly balls are falling in. This says that a pitcher can’t control that, but they can control how much they strike out people, how much they walk people, and how often they give up ground balls, line drives, and fly balls, and if these balls in play behaved the way they should, the pitcher should be getting this percentage of people out.

I will address the flaws I have found with it. As much as getting people out is important, sometimes what happens in the plate appearances that don’t end in outs are almost as important. This only deals in batting average regarding balls in play, but wOBA is very important too. Fly balls are more likely to be outs than ground balls, but the wOBA on fly balls is over 100 points higher. Additionally, I’d prefer instead of ground balls, line drives, fly balls, to use soft contact, medium contact, hard contact, because that is a truer test of pitcher skill, however, I did not have this data at my disposal as far as league averages on what the batting average is for soft contact, medium contact, hard contact (if someone does, please contact me like I said). So what I have for now will do and this batted ball data is still a good measure. I set out to calculate what percentage of batters a pitcher should be getting out, and that is exactly what I found out. So while it’s not perfect, it has its use, and it’s something to build on.



And build on I did. While the out percentage is nice, it doesn’t give us a measure like ERA or FIP or xFIP, that tells us how many runs a pitcher should be giving up. So using the data I used to calculate exOUT%, I present to you exRP27 (expected runs per 27 outs, a stupid name for a hopefully not stupid stat).

The basis for this stat is this data from FanGraphs, “Line drives are death to pitchers, while ground balls are the best for a pitcher. In numerical terms, line drives produce 1.26 runs/out, fly balls produce 0.13 R/O, and ground balls produce only 0.05 R/O.” (I don’t know how this was calculated, or when it is accurate for, but this is what I got). We don’t know this for soft contact, medium contact, hard contact, so again I’m sticking with ground balls, line drives, and fly balls. 

All right, so what I am going to do using this stat and the pitcher’s K%, BB%, GB%, LD%, and FB% is see how many runs the pitcher should be allowing over 27 outs, and then adjust it to get it on a scale similar to ERA, FIP, and xFIP.

Keeping Clayton Kershaw as our example, let’s take a look.

Kershaw’s K% is 32.3 — we’re multiplying this by 27 (for outs in a game), and we get 8.721 K’s, so 0 runs so far because a K will never produce a run

Now GB%. His exgbOUT% is 25.23354424, multiply this by 27 and we get 6.8 (ish, final number will be exact via the Excel doc). Multiply this by .05 (the runs per GB out he gets) and we get .34 runs.

LD%- his exldOUT% is 4.193784, multiply by 27 and get 1.13232168, and multiply this by 1.26 for LD runs/out and we get 1.43 runs

His exfbOUT% is 13.3181291, now multiply by 27 get 3.6 and then that by .13 and you get .47 runs

Add up all these exRUNS and Kershaw’s total is 2.24. However, we can’t stop here because the number of outs he’s recorded is only 20.3 (8.7+6.8+1.1+3.6) approximately. 20.3 is the rounded up total. So get this 20.3 (or whatever the pitcher’s exOUTS is) up to 27  by multiplying by whatever it takes, and then multiply his exRUNS by this same number. For Kershaw you end up with 2.97 exRP27. The league average would be 3.78. Last year’s average ERA/FIP/xFIP was 3.74, but when I adjust everything to that, everyone’s exRP27 just goes down slightly (Kershaw’s from 2.97 to 2.94), but I want it to be on a more realistic scale where everyone’s totals are lower and a really good exRP27 is comparable to a really good FIP, like in the low 2s. 

So I don’t know what the statistic’s correct way is, but here is what I did to make it work. I calculated what his “ERA” would be using by multiplying his exRUNS by 9 and then dividing that by his exOUTS. His was .99, the league average was 1.26. I then did .99/1.26 to get .78 or so, I then multiplied that by his exRP27 and got 2.34. I felt like this was more realistic and in line with his ERA/FIP/xFIP. Obviously, can’t be the same because they measure different things, but just got in in the area. And the same is done for all pitchers. Obviously, not everyone gets multiplied by .78 of course. The league average remains 3.78, between last season and this season’s average for ERA/FIP/xFIP.

Here is the leaderboard for that (S column):

View post on

 I really like this stat a lot, and feel like it does what I wanted to accomplish: figure out how many runs a pitcher should allow per 27 outs given his K%, BB%, GB%, LD%, FB%, and the notion that balls in play will behave the way they normally do, as anything else is likely luck and not indicative of the pitcher’s performance.

I look at Sonny Gray as someone this stat is perfect for. His ERA is outstanding at 2.04, but his FIP is 3.00, his xFIP is 3.47 and his SIERA is 3.50. The problem is, at least with FIP and xFIP for sure, is that they ignore what happens when the ball is in play. He doesn’t strike out too many people, he has a good BB% but not spectacular, and he’s given up 10 home runs, a fair amount, so this hurts his FIP and whatnot. However, instead of saying “well he will regress, look at his FIP/xFIP/SIERA” this looks at why he’s having this success, and it has to do with the balls in play, which is getting ignored. Gray’s LD% is just 14.6! That is really good! Second best of the 90 pitchers I did this for. And his GB% is 54%, 9th best, also really good. The pitcher does have control over the type of contact he allows, and the fact that Gray is producing a ton of ground balls, and very few line drives, is why he’s been so successful. His 2.34 exRP27 suggests that he has not been as good as his 2.04 ERA suggests, but he’s not as far off as the other stats suggest. 

Obviously exRP27 is far from perfect, and is in no way supposed to replace FIP/xFIP/SIERA, but it is something to look at with them. I am a big believes in aggregation, so I think that averaging some combination of these 4 stats together or them all, is an even better way to evaluate a pitcher. We’ve got more data than ever, so it makes sense to use it, exRP27 and exOUT% are just more examples of utilizing this data to help better evaluate pitchers.  

I hope you guys enjoyed. Any feedback please comment or contact me. Next I will be looking at exWOBA against for pitchers using similar data, and exWOBA for batters using the data but for hitters.

Carrasco’s New Deal, and Why the Yankees Should Do the Same with Pineda

I’m about to drop a cold, hard truth-bomb…

I’m not a professional general manager.


I know your mind just exploded, but it’s true.

Anyway, what I’m saying is that if I were a general manager (again I’m really not), I would hand out a lot more contracts like the one the Cleveland Indians just gave Carlos Carrasco, 28. For those of you not familiar, they agreed on a 4 year- $22 million contract. That shakes out to $5.5 million per year.

Now Carrasco is far from a sure thing as a top of the rotation guy, but he did have an encouraging season last year. He had 9.4 K/9 and an impressive 4.83 K/BB. Also, his FIP was just 2.44, suggesting that it wasn’t a fluky season, but a sign of things to come.

While we should expect some regression to the mean with Carrasco and can’t expect him to post a 2.55 ERA over the next four seasons, all the metrics suggest that Carrasco has what it takes to be a very good starting pitcher.

This isn’t an (article? blog post? stupid collection of words?) about Carrasco, though, it’s more about the type of contract he was given. We’ve seen it before, a player in his twenties being locked up to a long-term, rather low-per-year deal. Andrew Friedman was notorious for doing this with the Rays. For example, he locked up Chris Archer to a 6 year/$25 million deal when Archer was 25. Matt Moore got a 5 year/$14 million extension at just 22. Most notably, he gave Evan Longoria a 6 year/$17.5 million extension with an upside of $44.5 million over 9. These are good deals. Andrew Friedman is smart, so Andrew Friedman made these deals. (Logic!)

Why are they smart? Well, for a small-market team like Tampa, the deals allow them to maintain their homegrown stars for a longer time and at a relatively low average salary. For a big market team like the Yankees, these deals also make sense because if the player fails, it’s not a big deal to just eat the money they owe him. For example, if the Yankees decided to give Michael Pineda an extension in the range of 4 year/$30 million (give or take x million, I can not stress enough how bad I am at projecting contracts), to kick in starting in the 2016 season, I think that would be a really smart move, for both the Yankees and Pineda.

Pineda, when not injured or poorly concealing pine tar, has been a really good pitcher. I don’t want to bore you with numbers, just kidding I do. He has a lifetime FIP of 3.16 and a 3.78 K/BB ratio in 253.1 innings. Last year, he was filthy, posting a 2.61 SIERA, 2.71 FIP, and 8.43 (!) K/BB ratio. So, yeah, when he’s not a bonehead or hurt, he’s pretty freaking good. I recognize the inherent risk he carries, but (please don’t yell at me) he has shown flashes of a pitcher who can command $100 million when he hits free agency. Having a guy with that much upside and skill through his age-30 season at just 7 to 8 million dollars per year is really a bargain. If it doesn’t work out, they’re the Yankees and can afford to eat the money. It’s not like its a huge, burdensome contract.

The deals also make sense for the players, however. Look at Carrasco, first. Last season was the first in which he did not spend any time in the minors. Sure, the way he pitched suggested that if he continued like that and hit free agency eventually, he could be taking home a big contract, but when you have had just one, albeit good, season in the majors and you are offered $22 million, you probably take it.

Same goes for Pineda. He started 28 games in 2011, then missed two full season with injuries, and only started 13 last year. Sure, he’s looked awesome, but if you were a guy with his background of injuries and uncertainty, and you were offered $30 million, I imagine you take it. The deal also allows him to hit free agency when he’s 30/31 and, if he pitches well enough, get that huge contract.

So what have we learned:
1) I’m not a general manager
2) Long term/low AAV extensions can benefit both the teams and players
3) More contracts like this should happen