And an important note for paying subscribers only after the signature below.
WHAT GIVES?
Hi Peter,
The bid price for Alex Cobb ($3) seems low when comparing his projections to those of Reid Detmers ($10). Comparable IP, Wins, Ks, and WHIP. Cobb has a significantly lower projected ERA. I know the bids incorporate your general feelings about the players and how they are reflected in a maximum bid, but the difference between them seems too high. The discrepancy isn't as great, but Blake Snell at $7 has similar projections to Nick Lodolo at $11. Lodolo has more upside and Snell struggles to throw 5 IP a start, so I get that he has an edge, but again they are too close to have a 50+% difference in bid prices.
“What’s Up With That?”
Hey,
Thanks for asking. The distinctions between bid prices and projections are important, as is a discussion of what a projection is.
First off, the projections start out with a mechanical calculation for each player based on his past performance and age, with adjustments made based on Statcast data to compensate for what appears to have been good or bad luck. This works pretty well for players with a few years of track record.
For players with only a year or less of service, the mechanical projection can get weird. So these are the players I first prioritize with manual adjustments. These are my best attempt to incorporate common sense information to correct whatever skewness there is, and also to nudge these projections in the direction I think the player’s career is going to go.
I don’t go too far with this because research and my own experience show that the more we deviate from the norm the worse the entire projection set performs, though projection systems that include human judgment tend to do a little better than strictly mechanical systems if you look at the most active players.
Photo: Alex Cobb in 2011 with the Durham Bulls by Giants27
Here are the projections for Cobb and Detmers:
Alex Cobb: 31 starts, 166 innings, 9 wins, 3.60 ERA, 1.32 WHIP.
Reid Detmers: 30 starts, 166 innings, 10 wins, 3.58 ERA, 1.21 WHIP.
Last year, Detmers’ first, he earned $1. Cobb in recent years has earned -$4, $1, and $2.
Significantly, Cobb is 36 years old this year. Detmers will turn 24 in July.
The last three years Cobb has had a 3.84 ERA (xERA 3.15), struck out 8.75 per 9, and walked 2.86 per 9. So, the projection pretty much matches what he’s done.
In his rookie season, Detmers had a 3.77 ERA (xERA 4.05), struck out 9.51 per 9 and walked 3.21 per 9. So, his projection, based on limited past performance, is a little optimistic but not crazily so. And as you note, pretty darn similar to Cobbs’.
So, why would I pay $10 for Detmers and only $3 for Cobb?
Because Cobb has proven his general level of performance, and given his age it would be crazy to expect him to go much beyond that. Meanwhile, Detmers performed creditably in his first major league season. He’s young, approaching the age when experience and growing confidence can click in, and has two effective breaking pitches. If he throws more strikes, and gets ahead in the count more, he could quickly become a solid mid-rotation starter, the sort of guy who might earn $25 in his best years. The sort of guy somebody is going to pay $10 for, hoping to catch lightning.
And they might catch it.
Would I pay $10 for Detmers? Personally, I try to stay away from somewhat speculative plays on pitchers who cost in the teens. Some of those work out, but as a group they have the worst track record of success. On top of which, they cost real money, making it hard to stick to a $55-$65 pitching budget.
Will someone pay $10 for Alex Cobb? I don’t think so. Maybe someone will pay $5. That’s the difference.
The situation with Blake Snell and Nick Lodolo is similar.
Snell was pretty effective last year, but for the second year in a row he ended up with 128 IPs. Health is a skill, someone said, and while the uptick in innings in the second half last year and the dramatic improvement in his walk rate allowed argue for continued success this year, I’d be nervous getting too far out in front of that.
But I’ll admit that after looking more closely at Snell’s second half, which featured a 12.26 K/9 and 2.31 BB/9, I’m inclined to boost his bid price up to $10, nervously.
Lodolo, on the other hand, showed last year he can pitch in the major leagues. Like Detmers, he lives with his breaking pitches, and if he gains just a little more control he could take a big step forward, too.
To recap: The projections are rough approximations meant to reflect the mean evaluation of a hitter’s or pitcher’s true talent. What actually happens will land in a range of outcomes not quite as extreme as Everything Everywhere All at Once.
The bid price is where I think I would stop bidding on that player, all else being equal. It’s a bet. If I want the player I try to set the bid price at the place just above where I think the bidding will stop. If I don’t want the player I set the price a dollar or two below where I think the bidding will stop.
For these prices, which others are using, my emphasis is on where the bidding will stop, not so much on who I want. I would hope you would input your own preferences into the list.
All auctions are, obviously, different, so these are rough approximations, but as the auction season starts the various expert leagues become test cases for these, and I shift them up or down while keeping the overall budget at $3120 for the league.
Hope this helps. Thanks for asking.
HOUSEKEEPING
Thanks for reading. If you would like to learn more about becoming a paying subscriber, please click the link.
One of the perks paying for your subscription gets you is access to the Rotoman Pitcher and Hitter Sheet, which contains player prices for roto leagues, historical Cost and Earning data, games played by position for major leaguers, all of the player comments that have been published in the newsletter, plus others, and a growing set of player projections.
All of that is hidden away behind the paywall. Thanks for considering paying.
Keep reading with a 7-day free trial
Subscribe to Rotoman's Guide to keep reading this post and get 7 days of free access to the full post archives.