## Importance of Category Scarcity in Fantasy Baseball

We hear about position scarcity all the time, but category scarcity also plays a role in valuing players. In 2000, 47 players hit at least 30 HR (hmm, wonder why?) as compared to just 18 players in 2010. Mark Reynolds hit 32 HR last year and tied for 10th in baseball. Many fantasy owners continued to start Reynolds every day despite his sub-Mendoza .198 average because his power was so valuable. Had Reynolds hit 32 HR with a .198 average back in 2000, he would have been riding the digital pine. Power wasn’t at a premium back then.

And that’s category scarcity in a nutshell. In fact, position scarcity is really just a function of category scarcity. Shortstop is only considered shallow because there are so few players who can contribute across the board. A quick look at any shortstop rankings shows how rapidly talent plummets at the position.

The following graphs provide visual representations of category scarcity in 2010. They show how many players at each position posted at least league average totals in the five standard offensive categories (AVG, R, HR, RBI, SB). League average was calculated using all players who made at least 400 plate appearances. For reference, the league averages were as follows: .271 AVG, 70 R, 16 HR, 68 RBI and 10 SB (looks a lot like Justin Upton’s 2010 season, huh?).

*Note: All data labels have three parts: position, number of players that qualified and percentage of all players that qualified. On the graph right below, “OF, 26, 14%” means there were 26 OF with at least 400 PA, which represented 14 percent of all batters with at least 400 PA.*

Before we get into individual stats, we’ll take a look at the player pool we’re drawing from. Last year, 205 players had at least 400 PA. Because some players play multiple positions, the total pool of players increased to 239. Kevin Youkilis qualifies at both first base and third base, so we want to make sure we reflect this in the data. Now, to avoid the sheer quantity of outfielder from skewing the data, the total number of outfielders was always divided by three (since there are three outfielders for one of every other position). This put its totals on par with each other position and brought the total adjusted player pool to 182.

This graph shows the number of players who batted at least .271 last season, the league average. First base and shortstop accounted for 21 and 19 percent of all batters that reached the mark, respectively. These percentages are identical to their respective percentages of players with at least 400 PA, which tells us these positions were exactly average in terms of batting average scarcity last season. Second basemen accounted for 22 percent of all players who hit at least .271, a four point increase compared to the total number of second basemen who accrued at least 400 PA (18%), which tells us that batting average was plentiful at this position. Outfield also showed an increase (16% to 18%) while catcher and third base showed the only decreases. This means batting average was tougher to come by at both catcher and third base, and players from those positions who contributed in this category gained significant value.

By now you get the drill. Only one catcher (Joe Mauer) scored at least 70 runs last year, which means his contributions in this category were unmatched at his position and he derived a great deal of his value from his ability to score runs. Conversely, large increases were seen at first base and, most of all, outfield. These two positions had the greatest percentage of their players who qualified with at least 400 AB beat the league average of 70 runs. Second base and third base saw negligible changes of just one percentage point, and shortstop saw a three-point drop from 19 percent to 16 percent.

No shockers here. First base dominated home runs as we expected, which means it was easy to find a first baseman who beat the league average in homers. It’s also no surprise to see power was most scarce at shortstop (19% to 8%) and second base (18% to 11%). Power was relatively available at third base (18% to 21%) and catcher saw a rather negligible change.

Because RBI are largely a function of home runs, the two graphs are very similar. Shortstop and second base were once again the two hardest positions to find RBI whereas first base was once again the easiest. Outfield showed a slight increase from 14 percent to 16 percent as compared to all players with at least 400 AB but dropped when compared to all batters with at least 16 home runs. This tells us you could find home run production in the outfield, but RBI production was a little harder to come by. As we’d expect, third basemen saw a slight increase compared to all third basemen with at least 16 home runs and an overall increase compared to all third basemen with 400 PA. Catcher showed relatively no change.

Steals were virtually non-existent at catcher (seriously?!) and only a few hitters at first base managed to reach the league average (10). Not surprisingly, two of these batters, Albert Pujols and Joey Votto, are often considered top-tier fantasy options. While those two would likely be top tier first basemen regardless of their stolen base totals, they are widely recognized as noticeably better than the other first basemen (Miguel Cabrera, aside) as a result of their well-roundedness.

There were plenty of stolen bases available at shortstop, so a player at this position does not gain much value from stealing a lot of bases. As we saw, power at this position is most scarce, and that’s a big reason why Troy Tulowitzki is considered a mid-first round pick this season even though his cumulative 2010 stats were only 18th best among batters on ESPN’s Player Rater. Steals were also prevalent among outfielders and rather scarce amongst third basemen. Personally, this is why I have David Wright ranked ahead of Evan Longoria. With the lack of speed and batting average at third base, I find those two attributes most valuable at the position, and they happen to be the two places Wright excels over Longoria.

Remember, these graphs show how 2010 played out, but they can be used as valuable predictors for 2011. If you expect there to be more home run production from catchers this season, you can estimate an increase in their percentage from the graph above. Whatever you do, make sure you consider the importance of category scarcity when putting together your own position rankings.

Print This Post

This is great analysis. I agree that category scarcity is important, but I’m not sure I’d come to the same conclusion w/r/t the positional ranking adjustments (Longoria/Wright).

Because I do mostly auction drafts, I’d be inclined to use the info that 3B is weak in SB and AVG as evidence that those categories should be targeted when drafting other positions. Why pay for steals and average at 3B when I can get them cheaper at SS, OF, etc.? If I draft a SS and a 3B that combine to produce 30 SB, 40 HR, and .280 AVG, it seems I can get those cheaper by drafting the bulk of the speed at SS, the HR at 3B, the .260 AVG at 3B and the .300 AVG at SS.

This seems just as good a place to ask as any>

Never been in a roto/fantasy baseball league. Is there fantasy baseball site that uses a scoring system more in tune with oWAR accumulations? One that is not so dependent on those incidental things like runs and RBI’s but scores weighed more towards hits, walks, XBH’s, SB’s, K’s, WHIP, etc…

Thanks

74 outfielders had 400+ pa in 2010. 74/3<26

Only 24 2B had 400+ pa, not 33.

What am I missing?

Ben:

I can’t say with certainty, but you’re probably using stricter position eligibility standards.

Strange not to see Alcides Escobar among the top 30 fantasy shortstops on your rankings. He’s a fairly decent candidate for .271+ 20 SB+.

this is an extremely blunt instrument — much more precision is required to make this analysis useful. In other words, scarcity is definitely important, but the real question is exactly how important. For example, you state conclusorily that you rate Wright over Longoria because of scarcity. But given a projection for each player, you can actually calculate their value in light of scarcity. are you aware that you can use maths for this?

Also, who still uses 5×5 categories? Get yourself a points league.

Thanks for all the comments, everyone.

@Marver – We’re compiling our top 100 and top 200 rankings this weekend while making any adjustments we deem necessary at the individual positions due to recent trades and acquisitions. While we’re not high on Escobar since Kansas City has a less than good track record of developing hitting prospects in recent years, he will be debuting in the mid-20s.

@reader – You’re correct that category or stat scarcity alone isn’t enough to justify taking Wright over Longoria, but many people have these guys ranked 1a and 1b. In our podcast this week we touched on this and I referred to choosing between them as more of a coin flip than we see between the top 2 at any other position. Since I view them as statistically equivalent, although not identical because of the categories in which they contribute, category scarcity helps me lean toward Wright. And I still use a 5×5 league, both head-to-head and roto! Why points leagues? Ha I find them so boring.

“Never been in a roto/fantasy baseball league. Is there fantasy baseball site that uses a scoring system more in tune with oWAR accumulations? One that is not so dependent on those incidental things like runs and RBI’s but scores weighed more towards hits, walks, XBH’s, SB’s, K’s, WHIP, etc…”

Yahoo Baseball will allow you to set up a custom league with a point system. You get to choose 14 or 15 stats for batting and a similar number for pitching. Then you get to wait them as you choose (either plus or minus and can use decimal values). I have been running such a league for years. I set the stats for hitters, at the beginning, using partial differential equations on the Bill James Runs Created equation. I scaled everything so the superstars will score about 1000 points in a season. I set up pitchers in a similar manner (no differential equations though) to make the best ones also come out around 1000 points. Over the years I have tweaked it so that the top point scorers pretty much match the list of top vote getters for MVP and Cy Young. In “Real” baseball SB are not so valuable as in 5×5 roto, but I make middle infielders match actual value by giving points for Assists.

You could do something similar. If you’d like to use my system as a starting point, I’d be happy to send it to you. If you’d like to just join a league where guys you think are good in reality are just as good in fantasy, I probably have an opening this year. Send me an e-mail.

jimdetry@yahoo.com

I disagree with your assertion that position scarcity is simply a function of category scarcity on a number of levels:

(1) In any league where the active roster isn’t the same or exact multiple of the starting positions, position scarcity exists independent of category scarcity. In a traditional roto league with two cachers and five outfielders there is greater scarcity in those position than in the infield where there are three corners and three middle infielders. In a 12-team A.L only league, for example, teams will select 24 catchers while there are only 14 full-time equivalent catchers in the real game; by comparison teams will select 36 corners when there are 28 full-time equivalents — there is a scarcity of catchers compared to infield corners.

(2) You have made a logical leap but haven’t laid out a basis for doing so. You establish that some categories are more scarce for certain positions — which is obvious to anyone who follows the game. Then you make the leap: That it is somehow more beneficial to select players strong in categories that are in short supply for that position. Why? Why is it beneficial to target a catcher who can steal bases rather than going after a catcher with some power while targeting steals at other positions? The answer is it’s not — at least not inherently. What is valuable is targeting categories in ways that the market undervalues —— specifically the marketplace of your league. If everyone goes after a catcher who steals bases you can bet the opposite will occur — that catcher will be over-valued. As an owner in a fantasy league you examine what has been under and over-valued in the past, project that as you prepare for your auction or draft and adjust as that auction or draft unfolds.