This past week, in the ottoneu Slack Community, (check this out if you are interested in joining, which you should be if you play ottoneu) there was a heated debate about paying for top RP, particularly in points leagues. One side was arguing that the elite RP put up a Pt/IP score that few other players can manage. That if you are willing to pay $35 more for Mike Trout than Ryan Braun (projected to have a 200 point gap between them), why not pay the same premium for Aroldis Chapman over a guy 200 points below him – Santiago Casilla?
The other side argued that spending $30+ on a relief pitcher would be crazy – there are always solid RP options on the wire, so why not just wait and see who will be the next Wade Davis?
Both sides are right.
Only 10 RP project to put up more than 8 pts/IP in FanGraphs points leagues and only 3 project to be over 8.5. That kind of production is incredibly valuable. And if you take the projections at face value, paying $30+ for one of those three is an easy choice. Those are almost sure things to produce that value – I am just as confident projecting one of those three to be the top RP in ottoneu as I am projecting Clayton Kershaw to be a top three SP.
There are two problems with this:
One is that there is so much uncertainty in RP projections, particularly outside that elite group. Pitching is inherently volatile and RP are dealing with tiny sample sizes. The best of the best are likely to perform at a very high level. I’d be just as confident betting that Chapman will be a top-three RP as I am betting on almost anyone else to be top-three at their position. But the lower the tier, the higher the volatility, and at RP this gets extreme. So the chances that a $3 RP outperform a $5 RP are greater than a $3 OF outperforming a $5 OF. And the result is that you should actually be less willing to pay retail price for that $5 RP. If you are splitting $8, the $5 OF and $3 RP is probably the better bet. But since everyone knows that is the better bet, demand for the $5 RP is low, meaning that his price has to come down as well (basic economics – supply is steady, demand moves down, price has to move with it).
The second is that there are enough fantasy players who don’t believe in paying for RP that the total dollars spent on RP will almost always be lower than the projections suggest it should be. The top 60 RP in the FG points projections linked to above are projected to cost $640 total – or about $50 per team. That just is not going to happen.
The combination of the two effects is that RP cost less than the projections suggest they should. The first says that even a rational owner should not be willing to pay full price for RP, because of the volatility in their valuations. The second says that some set of owners artificially deflate their own valuations of RP.
The impact is that, even though the top RP will likely produce $30 of value of more, and even though they are worth that much in the auction, they should not actually cost that much.
The difference is important. I am pretty confident saying that if you pay $30 for Kimbrel or Chapman, you will get good value. However, Cody Allen, Dellin Betances and Brett Cecil all project to create more than $15 in value and will all probably cost less. Why spend $30 on a $35 Kimbrel, when you can spend $30 or perhaps less on $50 worth of those three?
At the end of the day, there are two ways to view a players auction price – one is in relation to the value he’ll create; the other is in relation to your other options. Elite RP are worth $30 on the first measure, but not by the second.
For me, $20 is about my cap for an elite RP in points leagues. They are well worth $30, but because RP are hard to predict and generally underpriced, I simply refuse to pay retail – I’d rather grab 5 RP for $5 who I think are worth $15 each, rather than pay $30 each for the top guys.
Chad Young is a product manager at Amazon by day and a baseball writer (RotoGraphs, Let's Go Tribe), sports fan and digital enthusiast at all times. Follow him on Twitter @chadyoung.