## The Ideal Hitter/Pitching Mix in Auction Values

There’s been some talk about the ideal auction budget mix on twitter recently. Perhaps, if you click this link, you can see some of the conversation between Chris Liss, Mike Gianella, Steve Gardner, Peter Kreutzer, and Jeff Erickson upon which I was snooping. It’s a fairly complicated conversation, and far-reaching. Let’s jump in.

The crux of the argument was that your split between pitcher and hitter values matters a lot to your auction values. The aggregate split is around 70/30 right now, but that’s just established market truth — that does not mean that the market is using the most ideal split right now.

You can look backwards and try to derive it empirically. For example, our retro-active auction values that we are going through (you can look backwards position by position by using the positional rankings links on your right). If you sum up the above-replacement hitters, you get $1677.40, and the pitchers give you $1164.60. That’s a 59/41 split! That’s almost definitely because our calculations assumed 13 hitter lineup spots to nine pitcher ones, or a 59/41 split on the roster.

But if you think that pitchers are more volatile than hitters, then you’ll want to push that 59/41 split further towards the hitters. You’d rather draft safer hitters. We know, for example, that pitchers get hurt more often than hitters (by decimal points) and stay hurt longer (by many days). That sort of volatility is easily handled by projections — just project your pitchers for fewer innings to reflect the idea that any starting pitcher that pitched last year is 40% likely to hit the disabled list this year. So that’s a way to handle volatility.

We can try to look at the year-to-year volatility in the position. For example, how many top-100 pre-season hitters showed up in the end-of-season top 100? 33/68 (48.5%) if you use ADP and our end-of-season values. 16/31 (51.6%) pitchers. 14/25 (56%) starting pitchers. Hold on now. Perhaps the top pitchers aren’t as volatile as we assume.

But what if pitching stats themselves are less reliable? We know they are — Dan Szymborski confirmed today — but how much more volatile are they? Much more. Look at this post evaluating the projection systems in 2011 — the correlation on hitters projections to their results was on the order of .62, and not a single projection system did better than .46 for pitchers. So clearly there’s a reason to distrust pitching projections a bit more than hitter projections, as many advancements as we have made.

Perhaps if you think the projections for the best pitchers are more stable from year to year, you can do something more like a 60/40 split at the top of your pitcher values. There’s some evidence for that — part of the problem with projecting pitchers is that innings-pitched numbers are more volatile year to year than plate appearances. The top pitchers may have more stable year-to-year IP totals, perhaps because they can move down the pecking order at their position and still pitch — the second-best second baseman on a team gets a huge hit in PA, but the second-best starting pitcher pitches almost as much as the first-best (provided both are healthy).

But if you look at pitchers as a whole, it does seem to make sense to push the needle towards 70/30, as we have all done. How far, exactly, might take a little more research.

Print This Post

I’ve been working on a WAR valuation model for a mixed H2H league I play in (5×5 with a few different cats), using win-probability-added calculations for each player in each category. My math could be wrong, but so far I’m seeing that pitchers produced 54% of the value in 2013. The top 20 pitchers added more than 20% more value than the top 20 hitters. I’ll have to look at your retroactive auction values as you are seeing the opposite….

very interesting….I have written several pieces saying that more ace starters should be drafted in the early rounds now that pitching is becoming more dominant. Would love to see the results of your research.

Where would be a good place to write up my methodology?

Community research on this very site?

Monty-click on my name.

Did you use the same lineup split? What positions did you use?

Wouldn’t that mean you want to not draft pitchers high up? In terms of Auction Values.

The reason being if there are more dominant pitchers, and specifically in the 2-3 spots on a baseball team. Wouldn’t it be better to let players to let players overbid for elite pitchers, and the snap up the next tier in pitching. Just because then if they do get injured you are not out of that value. If they develop into an elite pitcher, you are gaining more value.

But the difficulty is predicting those top 20 pitchers, or even pitchers 20-40. Projecting either hitters or pitchers accurately is a challenge, but the return on the value of hitters is higher because you are more likely to correctly identify the top ones (i.e. as a group they perform closer to their projections than pitchers do). So when you say “the top 20 pitchers added more than 20% more value” I assume you are saying the top 20 as identified after the season. If you are saying the consensus preseason top 20 produced that value, then that is surprising. I would want to know if that’s true of the preseason 20-40 as well, and then would want to know if that’s true of seasons other than 2013.

Agreed on the projection challenge and injury risk, but my point is that my actual retroactive values place a little more weight on pitching (46/54). In the article above the opposite is observed (59/41).

I have only one H2H league, but I find over and over and over again that pitchers aren’t just worth more than hitters, they are worth way more.

Part of the reason is that the league uses a shallow roster for position players but has no innings pitched per week cap and 10 weekly moves. Many teams will use 9-11 position players and 15-20 pitchers in a given week.

I suspect H2H leagues in general usually have a different hitter/pitcher split than roto leagues.

In head to head leagues the other thing that makes pitchers much more volatile than hitters is that their games played each week leads to a much smaller sample when compared to hitters.

Which means one bad pitching start can ruin rate stats. This is especially true in standard 5×5 as there are usually two rate stats for pitchers and only one for hitters. That poor start likely also puts you behind in any counting stat because the IP in a bad start are likely to be low.

On the other hand, one bad game from a hitter may dampen the only rate stat, but it won’t kill you. In addition, that player plays again the next day (and likely 3-5 more times that week) and can accumulate stats in those appearances. This isn’t true at all for pitchers.

With pitchers statistics more volatile in general and less stable in a week to week setting, it’s no wonder the dollar goes to the hitter!

well, top 30 starters are more dominant now than in past 10 years, so volatility isn’t as bad. To me, Kershaw is a first round pick in any league right now. Offense is down, and number of 80 run/80 RBI guys are shrinking, as is # of 30 home run hitters. Stolen bases are down as well.

Fantasy owners drafting offense early when there is less and less of it. why?

Because there is less of it, lowering supply and raising demand?

but how do we explain using the same strategy when offense was at its peak then? drafting hitters first is not new and has been a common strategy ever since I started playing.

Now that pitching is more dominant, and a little less risky, why aren’t we drafting SPs earlier now?

If there is less offense (scarcity), then we should spend more on it at auction, correct?

Drafting is somewhat different based on feel, position runs, etc but I don’t understand what you mean by:

‘Fantasy owners drafting offense early when there is less and less of it. why?’

Pretty sure the offensive environment doesn’t matter. You’re still filling out the same ‘x’ number of lineup spots and ‘y’ number of pitching spots, so all that matters is the value above replacement.

What if you looked at teams that won their auction leagues (or placed in the top 3-4) and calculate how much they spent on their pitching/hitting in the draft. There would obviously be a lot of noise, with trades, free agents, etc., but over a large enough sample, teams that have a better team on draft day are more likely to finish near the top at the end of the season. This might take quite a bit of effort but it would be very interesting to see if there is a difference between budget allocation in teams that finish at the top or bottom of the league.

I have a very different take on this and posted the first half of an explanation this morning. The split depends on two issues: The amount of money spent in the auction on players who don’t create positive value, for one, and the amount of available positive value that is available outside of the auction.

These two dynamics explain why we pay a discount for pitchers overall, but don’t pay like crazy for the best pitchers (who are usually as reliable in relation to preseason expectations as the best hitters). More detail in the story at rotomansguide.com and the second part, which I’m working on now.

thanks Peter….will check it out.

I’d say the ‘amount of positive value available outside the auction’ is directly related to the fact that projection systems are bad at projecting pitchers, but I enjoyed the piece!

Thanks.

I’m also reluctant to pay for mid-tier pitching given that there are usually enough April/May surgers who can be picked up cheaply and provide just as much value. My 2013 pitching staff was a mixed bag of misfit toys (Corbin, Lynn, Cingrani, Salazar, Estrada, Haren, Jimenez, and Kluber), but for a combined salary of just $45 they held together pretty well.

Finding mid-season hitting off the waiver wire has been tougher. With pitchf/x and plate discipline stats for pitchers, I find it easier to assess a pitcher’s likelihood of continuing to perform strongly, given a small sample of current season data. With hitters, a surge in performance based on luck seems more difficult to differentiate from a true improvement given a short time period of data (April/May numbers).

I just posted Part 2. I’ll answer questions here or over there. Thanks for your thoughts.

Steamer projections have a “Reliability” category, which ranges from around 50% for closers up to 90% for elite stable hitters. I use that as a starting point for docking certain players who Steamer projects as unreliable. It’s eye opening in that like I see in the linked article, some elite SPs deserve to be paid full price for but some unproven hitters such as 2013 Cespedes (70% reliability for 2013) don’t end up being good buys at all.

I’m not impugning the Steamer reliability index at all, but to judge it is important to understand what the measurement constitutes as reliable. And what something like 70 percent means. I use a Reliability measurement that simply counts the number of datapoints and consistency of playing time. The less data or variations in PT flags those numbers for double checking. I’m not sure they measure the reliability of the player or the projection however.

For me, it comes down to the fact that pitchers are difficult to predict (more than injuries, it comes down to luck) and that there are always replacements available. These replacements consist of pitchers who are undervalued (dropped after being victimized due to bad luck) and pitchers from the minors who succeed with their opportunities (because another SP gets injured or demoted). This is true even in deep leagues. I prefer to spend $1 – $5 per pitcher and overstock on cheap pitchers. That way you can cycle through them without regret. You profit off of every lucky streak and don’t get hurt by bad luck. I get that people want an ace, but people have lost value on Halladay, Verlander, CC Sabathia, Strasburg, Lincecum. Unfortunately, at some point Kershaw will disappoint just like other first-round pitchers have.

Rather than focus only on the relative volatility of top 100 players, it’s helpful to consider the volatility of opportunity for pitchers who are either far down in ratings or prospects.

We know starting pitchers are more prone to missed time to injury. I suspect too that starting pitchers are also more likely to losing their spot in the rotation than starting hitters are to lost their spot in the lineup — a 5th starting spot can be a revolving door for many teams.

The net effect is that pitchers are more likely to gain a chance to start during the season than are hitters. This effect creates a pool of pitchers with projected negligible value who will have a chance to create positive value. Their availability depresses the value of pitchers projected to contribute.

The effect is amplified in a keeper league. A pool of low-priced, productive starting pitchers is created that is kept from year-to-year and this diminishes the total amount spent on pitching.

Thanks for this article on auction values. As wonderful as Fangraphs has been, the weight of article has been on draft leagues, reflecting the popularity of that format. I find auction leagues more interesting as they seem more like what real MLB teams must do, assigning value to each player, not just relative placement.

This is a great point. Focus on opportunity with pitchers. Last year in an AL-only I drafted a bunch of cheap pitchers by pairing cheap pitchers with upside on the same team. Pair a cheap pitcher on a team with the #6 on the team, so if they get hurt, you don’t lose. Dan Straily/Bartolo Colon, for example. Straily filled in for Colon while he finished his suspension, Colon came back and was great and Straily finished the year off great. If you make sure to spend very little on these players, the extra roster spot required won’t hurt you because you can drop one of them if needed without harming your team’s overall value.

A fundamental piece that I believe is missing in the above conversation is streaming. In a standard ESPN/Yahoo daily league it should be much, much easier to effectively stream pitchers than hitters. A starter who puts up below replacement-level stats across a full season can put up valuable stats in a start against the bottom of the league. Depending on league settings, on any given night there might be 5-10 available SPs starting against the bottom third offenses in the league, and park factors can also be taken into account. Further, with probables announced days in advance and websites set up to sort through upcoming match-ups it makes it that much easier to rely on streamers to rack up innings. More analysis on streaming would be helpful.