Updating Prospect Valuations

Over the last five years, a lot of good work has been done on valuing prospects. Victor Wang — now employed by the Indians — got the ball rolling in 2008, and his work was followed up by research from by Scott McKinney in 2011 and Kevin Creagh in 2012. Each piece was well done and is worth reading even now, especially if you’re interested in the various rates at which prospect tiers tend to go bust.

But we’re almost into 2015 now, and baseball revenues are exploding, making the valuations from even a few years ago look a little bit outdated. Just in the last year, we’ve seen three Cuban free agents — Jose Abreu, Rusney Castillo, and Yasmany Tomas — sign for around $70 million each, suggesting that teams are willing to pay significant prices for young talent with upside, even when there’s not an established Major League (or even minor league) track record from which to evaluate. We’re going to get an even better look at the market value of a premium prospect when Yoan Moncada is declared a free agent, and the bidding for his services is expected to reach $80 million once the taxes are accounted for, and that won’t even cover any of his future salaries; that’s just the cost of acquiring his rights.

So, helpfully, Kevin Creagh and Steve DiMiceli decided to update their study this week, adding in new data and using updated calculations for the price of a win based on recent inflation. Let’s look at what they found.

The bulk of the information is found in the table summarizing their research. I’m going to show the table below, but you really should read their entire article, given that it contains a walkthrough of the methodology and some further explanations of how they reached these calculations.

Tier	Number of Players	Avg. WAR	Surplus Value	% Less than 3 WAR	% Zero WAR or less
Hitters #1-10	53	15.6	$48.4M	13%	9%
Hitters #11-25	34	12.5	$38.3M	32%	9%
Hitters #26-50	86	6.8	$20.3M	50%	31%
Hitters #51-75	97	5	$14.5M	57%	44%
Hitters #76-100	96	4.1	$11.6M	65%	42%
Pitchers #1-10	18	13.1	$40.4M	6%	0%
Pitchers #11-25	47	8.1	$24.5M	45%	28%
Pitchers #26-50	77	6.3	$18.7M	42%	25%
Pitchers #51-75	94	3.4	$9.4M	70%	48%
Pitchers #76-100	105	3.5	$9.6M	67%	45%

The most obvious thing to note: elite prospects are super valuable for a reason. The best hitting prospects from 1994-2005 averaged +16 WAR during their team controlled years, and the best pitching prospects weren’t even that far behind, at +13 WAR over the same period; the #11-#25 hitting prospects almost matched that total as well. A top 25 hitting prospect or a top 10 pitching prospect regularly produced high levels of performance in their pre-arb and arbitration years.

And as much as everyone likes to bag on prospects for being unproven, the bust rates of these types of players was actually quite low; only about 1 in 10 produced nothing at the big league level. At the very top of the prospect charts, Baseball American did an excellent job of identifying players who could make an impact at the big league level, and the risk associated with these kinds of players is generally overstated. Elite prospects often turn into good big leaguers, and the rarely turn into nothing.

As a point of comparison, I looked at the leaderboard for the best under-32 pitchers from 2006-2008, to see how well they would perform in the six seasons from 2009-2014. The 10 best not-old pitchers in that three year window: Roy Halladay, Johan Santana, CC Sabathia, Dan Haren, Jake Peavy, Roy Oswalt, John Lackey, Brandon Webb, and then either Javier Vazquez and Josh Beckett if you’re using FIP based WAR or Cole Hamels and Carlos Zambrano if you’re using RA9 based WAR.

During this three year span, these aces averaged roughly +15 WAR, making them established five win pitchers with a significant track record of success. The age limit helps keep the list to just pitchers who reasonably should have been expected to have six years left in their big league careers. How’d they do over these past six seasons?

Halladay, Sabathia, and Hamels were awesome, producing at least +20 WAR by either FIP or RA9 versions of WAR. Haren was close, getting +20 WAR by FIP and +17 by RA9, and he threw the most innings of anyone in the group as well. But after those four, it’s a pretty big mixed bag, with some good performances, some bad performances, and a lot of injuries. Brandon Webb basically never pitched again. Carlos Zambrano and Johan Santana were okay when they were healthy, but weren’t healthy much. Oswalt had some good seasons and some bad seasons; ditto John Lackey.

All told, the top 10 pitchers by 2006-2008 FIP-WAR produced +14 WAR/+13 RA9 from 2009-2014. The 10 pitchers by RA9-WAR produced +15 WAR/+14 RA9, doing slightly better because Cole Hamels gets added to that group and pulls the average up a bit. Webb is the big bust in both samples, offering essentially the same 1-in-10 zero value return as the prospect group. Besides Webb, there weren’t any other players particularly close to zero value, so the elite pitchers offered a slightly higher floor than the prospects, but their overall average performance was actually a little bit worse than the best hitting prospects, and essentially even with the best pitching prospects.

That’s right; for all the talk about unproven risky prospects and established #1 starters, the average performance of six years of an elite pitching prospect lately has been in the same range as the average future performance of six years of one of the best pitchers in baseball. You can re-do the same calculations for other years, and the numbers will fluctuate a bit with age and stuff, but in general, once you’ve identified a “true ace”, you’re probably looking at something in the +15 WAR range over the next six years.

And Creagh/DiMiceli’s work suggests that elite prospects have been performing at similar levels. In other words, even without taking the massive cost differences into account, elite prospects and frontline starting pitchers seem to have about the same long-term values. If we redid this analysis to focus solely on the short-term, the aces would undoubtedly win out — the six year timeframe clearly creates a bias towards young players, which all prospects are — but it’s worth knowing that the long-term future of a prospect isn’t that different from the long-term future of a pitcher who has established himself as a legitimate Cy Young contender.

And, of course, Cy Young contenders cost a lot of money, while players in their pre-arb and arbitration years come at a significant discount. For the surplus value calculations, go through and read the full methodology of Creagh/DiMiceli’s work, as they do a good job of discounting future wins and attempting to account for arbitration payouts. The one thing that I’d add to that section is that there’s also a significant value in having an elite young player because of the chance to sign them to a long-term extension that goes beyond just the 0-6 years, and often create value for the franchise in what would have been free agent years. Once you factor the extension opportunity into the mix, I think the values listed by Creagh/DiMiceli could be a bit low, and teams are likely placing an even higher value on these types of players than the numbers stated in their article.

But even with that minor quibble, this is still excellent work, and worth referencing as a baseline for when teams make prospect-for-veteran trades. It’s never as simple as proven versus risky, and as the data shows, the best prospects in baseball are probably not as risky as many think, relative to established big league stars. Kudos to Creagh and DiMiceli for putting in the work to update the data, and helping to provide evidence of the fact that elite prospects are being highly coveted for a reason.