Thursday, March 1, 2007

Position Scarcity

Will there be position scarcity in Rotohog? I've attempted to answer that question by determining the project points that a 'replacement level' player at each position should accumulate. These are the players that figure to be mostly unowned, and available for little to nothing:

1. I took all the ZIPS projections and used those to calculate expected points for each hitter based on Rotohog's scoring system.

2. For 1B, 2B, 3B, SS, and C I determined the points value of the 13th player. These #s were: 1B-752 , 2B-610 , 3B-715 , SS-632 , C-473.

3. I did the same thing for the 37th outfielder, whose points value turned out to be 679.

4. Of the players I hadn't 'used up' already, I took the next twelve overall (for the 'utility' position). The 13th had a points value of 676.

5. Rotohog has three bench spots on the roster, which can be used for hitters or pitchers. If you assume that on average one of these will be a hitter, eliminating the next twelve hitters brings you to a points value of 653. If you assume that people will generally carry two hitters on the bench, you're down to 636.

This was definitely a pretty crude process. ZIPS projections don't capture a lot of known information about expected playing time. I was assigning positions by what was in the ZIPS data, not what position the players are classified by in Rotohog, and I'm sure there are other substantial 'errors' as well. That said, I think this process should paint a pretty clear picture of where the position scarcity lies in Rotohog.

There is a very slight scarcity issue at second base, but probably small enough that it can be ignored when assigning values to players. There is a much more substantial position scarcity premium for catchers however! The replacement level catchers that are unowned and can be picked up for almost nothing are going to average almost 150 points worse than at any other position!


ZZ said...

Forgive me for asking, but isn't 'the optimal strategy' clear, at least in outline? Shouldn't one pick the absolute best people possible at first (say Pujols, Reyes, Utley, Crawford, Santana) and then fill your roster with 'penny stocks' that can only 'go up'. You can, I gather, trade up as the season goes along. But there's no reason you would ever trade Pujols or Utley or Reyes (barring injury, etc.), because you couldn't do better at that position.

Maybe I'm missing something, but it seems like the only key is learning how to press the button quickly, and having a second and third choice ready to hand in each round.

I signed up, but frankly, Rotohog doesn't seem likely to hold my interest...

Alex said...

zz - Nope. You'll do ok that way, but not great. There's a lot more to it than this, but a much better strategy would be to look for the most obviously undervalued players and draft those first (whether they include the top players or not). Because of the open market for trading, you should always be able to get the Pujols, Reyes, Santana type players at a fair price the day after the draft...and you may be able to sell some of your undervalued guys for a quick profit.

ZZ said...

Interesting; unfortunately, unlike the stock market, there's a cap on how good your team can be (from the constraints on roster size and position). Given that, I don't see why anyone would trade Pujols, except for an exorbitant fee.

We've seen strategies evolve in this direction (bigger split between top and bottom salaries) to a degree in auction drafts as a historical trend - in the original 'Rotisserie' books, noone ever went for more than $30, this year we see Pujols and Santana close to $50 in standard leagues. The particular format of RotoHog (fixed pricing, small inflexible rosters, very shallow penetration of the overall player domain) will exacerbate the situation.

Obviously undervalued guys are good to have, but the $.50 Joel Piniero's of the world are even better - no risk/relatively high reward. If Utley is the best you could ever do at second, why not take him first (even if you have to downgrade from a $20 Alexis Rios to a $5 Nick Markakis). You get a full year of Utley stats, and don't have to worry about the rest of the world realizing that Markakis is really good.

Since quality players will not be at a premium (since it's only 12 team, and shallow) it will be very hard to trade for a 'blue chip' player. In a standard fantasy league you can offer 2 for 1's, but the shallower the league, the harder that is to do (since it's not hard to fill slots with solid starters).

Another factor is that, also unlike the stock market, there's no reason not to spend your initial $275. In a real market the goal is (mostly) to increase your ROI, and a good strategy would be to look for cheap, undervalued commodities. (It would be a fun challenge to play rotohog this way - only draft players under $5 and see what you could parlay that into by the end of the season). But since everyone has to invest $275 you should get some of the best players, since that's where you want to end up anyway.

Plus, unfortunately, there will be enough 'dumb money' in the market that it will be very easy to capitalize on injuries, etc, while folks aren't paying attention. Thus, it shouldn't be much harder to get from Hermida to Crawford than from Ordonez to Crawford over the course of the season.

I guess my contention is that (most of) the best players just are undervalued, because the ranking system as it stands doesn't take into account their unique role as the very best at their position, and because of the fixed pricing. If Pujols were $80 rather than $50, I'd probably feel otherwise (but I wouldn't be surprised to see him at $75 by June). But since anyone whose team is 'making money' will be gunning for him (and Ultey, Reyes, Santana, Mauer), there will be a significant scarcity.

Anyway, it's fun to think about; we'll see how it goes...

Aaron said...

I have one minor comment on your analysis. It's not logical for a position to have a value higher than the utility slot. After all, you can always play a position player in the utility slot. That affects the position value of 1B (752), 3B (715) and OF (679), all of which should be lowered to 676 to match the utility slot.

All of this ignores the bench, which makes things more complicated. A position slot can have a higher value than a bench slot. It's true you can always put a position player on the bench but then (duh!) you don't get his stats.