Wednesday, February 20, 2008

Revisiting Hitting/Pitching Splits

Eugene Freedman of Baseball Think Factory has been posting his astute comments about the results of the CBS Sportsline expert auction I participated in back on February 5. I've appreciated the compliments (and taken the jabs in good fun).

He said:
I think the league overall undervalued pitching....What was the league's split?
I may have posted this elsewhere, but I'm not sure so I'll post it again. The league had the "standard" $3120 budget for $260 a team. A total of $2163 was spent on hitting and $953 on pitching. This amounts to a 69.4/30.6 percent hitting/pitching split.

This is nearly in line with Alex Patton's 4x4 formulas, which calculate values off of a 70/30 percent split, or $182/$78 per team. I've never peeled the onion to determine if he uses the same 70/30 split for his 5x5 formulas. but if he's using the same principles he's using in 4x4, the split should be $175/$85, or 67.5/32.5. If you're interested in the mechanics as to why this is so, follow this link.

I've always taken Patton's advice for my own auction leagues and budgeted using a $175/$85 split. However, in the same article linked above, I noticed that:
LABR paid 31% of its budget for pitching, while Tout Wars paid 30.2%, or nearly $78 per team. Ironically, both LABR and Tout Wars use the 5x5 format so, in theory, they should actually be paying a little more, not a little less, or pitching.
This isn't a trivial matter. Evidence from three expert auctions in a row (including Sportsline) indicates a slight tilt toward spending even less for pitching than the $85 per team that many of us have been spending in our leagues.

And this returns me to the first part of Eugene's question:
I think the league overall undervalued pitching.
From a theoretical standpoint, this is correct. From a practical standpoint, the value of each league's pool of players is what is paid for them.

Unlike some valuation systems, the Patton $ valuation system derives its values from the closed pool of players purchased at auction. Theoretically, the average stats of the 168 hitters purchased across Rotisserie Leagues throughout the land are worth $2184 for 4x4 leagues and $2100 for 5x5 leagues. In practice, though, the stats are worth what your league paid for them.

Eugene's complaint is valid in a theoretical sense.

From a practical standpoint, though, the hitters will be worth what we paid for them and the pitchers will be worth what we paid for them. In a closed player population, the hitters cannot earn a penny more or a penny less than what we paid for them, and the same is true of the pitchers.

From an even more practical standpoint, our bid limits for our league should be in line to what our league typically spends for hitting versus what is spends for pitching. If my carryover league typically spends 75% of its money on hitting, I'd better budget 75% of my bids for hitting. The hitters might not "earn" that much. But they will be paid that much, and I had better adjust to my league's own reality.

4 comments:

Rodger A. Payne said...

Mike, I think part of the difference is related to the issue I raised previously -- roto leagues draft a very large portion of the hitting, but a much smaller portion of the pitching.

Conceivably, after a draft, there is very little hitting value to be acquired, but plenty of pitching value remains freely available. Scarcity is playing a big role in pricing.

In turn, I think my AL league's move to expand the number of pitchers and decrease the number of hitters is going to have an interesting effect on the auction.

On the one hand, offensive scarcity will go down and pitching scarcity will go up.

On the other hand, perceptually, the pitching talent being drafted is a bunch of $1 guys who would otherwise go undrafted. It is possible that teams might allocate even more money to hitting because ERA and WHIP is going to be even more fickle with another pitcher on a team to pull down averages.

All the undrafted hitters will be immediately taken in the reserve portion of the draft.

mike fenger said...

I
think that Mike is right in saying (essentially) that the practical overrides the theoretical -- whatever a given league spends on pitching vis-a-vis hitting is what need to inform its players' bid values. I also agree with what I think Rodger is saying -- that typically, leagues will pay more in hitting than they "ought" to, and rarely pay more for pitching than they "ought".

And, the appropriate advice is to move against the room -- the bargains are in the area that the other owners are undervaluing. Be the Billy Beane in your auction -- buy up what the market undervalues (while making sure that your team is competitive otherwise).

Anonymous said...

Good stuff MikeG. My next piece is going to be about the 30% pitching split and where it comes from.

I think too many people take it for granted and it impacts 5x5 pricing too much.

T.J. said...

In line with the "Zag while everyone zigs, zig while everyone zags" comments: If the theoretical "correct" split is 70/30, and your league goes 75/25, I'd do something like 73/27 or 72/28.