Valuing the draft (Part 1)by Victor Wang
January 13, 2009
A lot of great research has been done in the past few years on the draft. I highly recommend these two series. Also, Nate Silver has done research on the value of draft picks. It is Silver's article that is the inspiration for my first draft article, as I try to estimate the value a Type A free agent gives away through draft picks.
What I first did was create a draft database that contained all players drafted in the first 100 picks from 1990-1997. I only included players who signed. I recorded the player's position, whether he was drafted from college, junior college, or high school, and the WSAB he produced in his first six years of major league performance. The use of WSAB is one area that is new in draft research. The only previous studies that I have seen have used Baseball Prospectus' WARP statistic, which is actually being remade. If you've been following my prospect valuation articles you'll know that I picked a player's first six years because that is how long a team has control of a player before he reaches free agency. However, this assumption may need to be changed in the future.
The process of valuing draft picks is almost the exact same as the process of valuing prospects. What makes draft picks valuable is that you can pay the player a lot less than he is worth during the player's first six years. This is called surplus value and is the performance value of a player minus his actual salary. I've gone over the process before and I refer you to my article here for the exact process or you could also read my article in the 2009 THT Annual. The only major change when dealing with draft picks is that we need to account for a draft pick's signing bonus.
Value of Type A draft picks
The rules for compensation picks can get pretty complicated. For the initial process of valuing Type A draft picks, I'm going to try and keep the assumptions as simple as possible. First let's go over the draft pick compensation process. Let's assume a team offers a Type A free agent arbitration and the player gets signed by another team. We'll call a team that loses a Type A free agent the losing team and the team that signs him the winning team.
By having a Type A free agent get signed, the losing team gets a draft pick from the winning team and an additional supplemental draft pick. The supplemental pick is made during the supplemental round, which falls in between the first and second round. The order of supplemental picks is given out in reverse
order of standings. The order has to cycle through teams before a team can pick twice. The draft pick they get from the winning team depends on where the winning team is picking. If the winning team is picking anywhere between 1-15, they get the winning team's second round pick. If the winning team is picking anywhere between 16-30, the losing team gets the winning team's first round pick.
However, a team can sign more than one Type A free agent, like the New York Yankees this year. They signed three Type A free agents, who were all offered arbitration: C.C. Sabathia, Mark Teixeira, and A.J. Burnett. Since the Yankees only have one first round pick they can give away, only one losing team can get the Yankees' first round pick. This is again determined by the Elias Sports Bureau. The losing team that lost the highest rated free agent gets the Yankees' first round pick while the team that lost the second best free agent gets the Yankees' second pick. That means the team that lost the third rated free agent ends up getting the Yankees' third round pick.
It turns out that Teixeira was the highest rated free agent, meaning the Angels will end up getting the Yankees' first round pick. The Brewers get the Yankees' second round pick for Sabathia and the Blue Jays end up having to settle for the Yankees' third round pick.
Now, events like this are typically rare. Calculating the probabilities of stuff like this happening would just add to the complexity of trying to value Type A draft picks. To keep things simple for now, I'm going to find the values for three tiers:
Tier 1: Picks 16-30, a losing team gets a pick in this tier if the winning team's first round pick falls in this range.
Tier 2: Picks 31-45, this is the equivalent of a supplemental pick. A team always gets a pick in this tier for a Type A free agent.
Tier 3: Picks 46-60, a losing team gets a pick in this tier if the winning team's first round picks fall in the 1-15 range.
To find the value of a draft pick in this tier, I first found the average WSAB that a pick in each tier produced. Then I estimated the typical surplus value that a player would generate with that average production. Note that these surplus values do not account for signing bonus cost. I found some very interesting results:
Tier 1 Surplus Value: $6.51 million
Tier 2 Surplus Value: $1.17 million
Tier 3 Surplus Value: $3.43 million
Interestingly enough, the average third tier pick has generated $2 million more in surplus value than the average second tier pick. My initial thoughts were that these results were caused by small sample size problems. However, each tier has about 120 players in the sample. This is about the same as my prospect samples, and there don't seem to be any problems with those samples. It could also because teams were scouting and drafting inefficiently during the 1990-1997 period or that certain draft picks were running into bad luck.
After thinking about this for a little bit, my belief is that the superior performance from third tier players is mostly due to noise. Projecting draft prospects is a lot more difficult than projecting regular prospects for a few reasons. Due to this and noise, third tier players happened to outperform second tier players from 1990-1997. However, let's say that their was an inefficiency and teams were underrating third tier players and overrating second tier players for some reason. Theoretically, teams would eventually be able to detect this inefficiency and start drafting third tier players in a second tier spot and vice versa.
Because of these reasons, I think that when we're trying to project draft pick value, Tier 3 value tells us more about Tier 2 value than past Tier 2 value. In fact, I would say that when assigning tier value, we should switch two and three around so we end up with this:
Tier 1 Surplus Value: $6.51 million
Tier 2 Surplus Value: $3.43 million
Tier 3 Surplus Value: $1.17 million
I feel that this will be more accurate in projecting the value of future draft picks. Next I subtracted signing bonus cost by averaging slots given out by the MLB for each tier and subtracting them from surplus value. I ended up getting this:
Tier 1 Surplus Value: $5.23 million
Tier 2 Surplus Value: $2.63 million
Tier 3 Surplus Value: $0.76 million
You can see that going from a Tier 1 pick to a Tier 3 pick is pretty costly. It's also clear that the exact value of Type A picks depends a lot on the chances of getting a Tier 1 pick or a Tier 3 pick. Here is how the value changes if we vary the chances of getting a Tier 1 pick and a Tier 3 pick:
Chances Tier 1 Chances Tier 3 Surplus Value 1 0 7.86 0.75 0.25 6.7425 0.5 0.5 5.625 0.25 0.75 4.5075 0 1 3.39
However, these are the values when you just vary the odds of a Tier 1 and Tier 3 pick. Teams also need to account for the chances of a Yankees-type situation. Also, teams need to factor for the chances they decide not to offer arbitration and the chances a player accepts arbitration when offered.
If teams decide to put a a value on draft picks they may receive for a Type A free agent, there are a lot of probabilities they'll need to calculate. Given the uncertainty of off season events, it can be rather difficult in estimating these probabilities. Because of this, I feel that the most reasonable projection for the value of Type A draft picks would be something between $3-5 million. I'll make sure to note this update in any future trade valuations. Up next I'll be comparing the performance of high school and college draft picks.
References and Resources
Vince Gennaro has also written about the economics of drafting and player development.
Victor Wang's work on OPS has been featured in SABR's By the Numbers magazine, and was the 2007 recipient of SABR's Jack Kavanagh Memorial Youth Baseball Research Award. He can be reached via email here.