The Sports Stock Marketby John Beamer
April 23, 2007
Every year a few cool and innovative baseball websites pop up. 2007 is no exception and one such site is PROTRADE, which claims to be the first sports stock market—and a darn fine job it does of it too.
Regular readers of The Hardball Times will know that I’m hooked on prediction markets. A couple of weeks ago I looked at what the markets thought about THT division standings, and last week and I talked about Tradesports, which acts as a futures market for baseball. Assuming I haven’t sated your appetite yet, this week I’m going to twaddle on about PROTRADE.
What is PROTRADE?
As it says on the tin, PROTRADE is a platform for trading sports players and teams (think stock market). The firm is well backed with venture capital financing and seems to have a few heavy hitters piling in some cash: Billy Beane, Troy Aikmen and Brent Jones are all cited as investors.
This is what the folks at PROTRADE say about themselves:
PROTRADE is a next generation fantasy sports experience where a community of passionate sports fans trade players and teams like stocks. Our virtual sports stock market helps capture the wisdom of these sports fans by enabling them to display their reactions and generate a market response to every event in sports news (every at-bat, every drive, every rumor, and every injury report can be factored into the value of an athlete or team).
That sounds remarkably like the futures markets we looked at last week. However, there are some subtle differences, and actually the two markets co-exist nicely. The main difference is that a futures market measures definite outcomes whereas a stock market only measures absolute value or performance. Want to know if A-Rod will win the AL MVP? Only a futures market can tell you. However, if you're more interested in the relative value of, say, A-Rod and David Ortiz then a stock market is the place to look.
How PROTRADE Works
It is quite simple really. The market functions as a fantasy league with a trading mechanism bolted on. At the start of the season PROTRADE sets the price for all players based on its internal projections and fantasy scoring system. Bidding then opens and from that point player prices move depending on market whims. If traders think that a particular player is undervalued (ie, people expect him to outperform) his stock will be bid up.
PROTADE allocates points depending on how hitters and pitchers perform on an array of count stats. Hitters are allocated points as follows (1 point = $1):
Hitting Stats Points Hit 1 Out -0.38 Run 0.74 RBI 0.68 HR 1.67 SB 1.75
While pitchers are allocated points for:
Pitching Stats Points Win 4.68 Strikeout 0.45 Save 2.15 Earned Run -0.98 Hit or Walk -0.53 Out 0.38
The theory behind choosing these metrics is that they are associated with traditional 5x5 rotisserie leagues—with batter and pitcher outs used to determine batting average, WHIP and ERA.
So, if you think that Alex Rodriguez, for example, will get 166 hits, 406 outs, 113 runs, 121 RBIs, 35 home runs and 15 stolen bases then his value is $262. Those were his stats last year—given that he is currently doing a fine impersonation of David Ortiz’s doppelganger, the market, at $363, is pricing in considerable upside.
PROTRADE attempts to mirror the stock market in other ways. To help traders make investment decisions, it places buy/sell ratings on players that are perceived to be under or over priced respectively (based on last year's performance). It is also possible to execute more complex trades. Want to short sell, or cover a position? PROTRADE lets you do it. For those with a keen interest in trading and commerce it is a fun (and harmless) introduction to the machinations of markets.
PROTRADE is squarely aimed at the fantasy market, and as such that limits its usefulness for more mainstream baseball analysis. While the market may reflect the fantasy value of a player that doesn’t necessarily mean it reflects the true win contribution of different hitters and pitchers.
For instance, take pitcher stats, which include measures such as wins, saves and earned runs. We know that these metrics don't necessarily reflect true talent. For instance, who do you think is the better player: Rafael Soriano or Bob Wickman? PROTRADE would unfairly say Wickman (by two times) by virtue of his accumulating more saves. The same is true of hitters as middle order batters are rewarded for racking up RBIs, which is tougher for the leadoff guys to do. Also look at the relative value of different offensive events: a home run is worth 1.67 and a stolen base is worth 1.75. That egregiously overvalues the stolen base, and there is no penalty for caught stealing.
However, the folks at PROTRADE are running a business. Let’s face it, the market for mainstream baseball valuation is probably relatively small, while the market for fantasy is huge. They’ve got to make money, and targeting the fantasy audience is a great way to do it. This needn't be a problem for other analysts as we can still get a fair idea of the relative value of different players and teams. If PROTRADE catches on and becomes uber-profitable, then hopefully it will diversify into other, more relevant, markets.
Anyway, it is what it is, so let’s dive into some of the great data available.
Here is a list of the 10 priciest pitchers (as of Saturday, April 21):
Stock Price J. Santana $265.90 R. Halladay $243.74 R. Oswalt $241.09 J. Peavy $240.21 B. Webb $231.94 B. Sheets $225.02 C. Zambrano $221.38 J. Bonderman $209.18 C. Sabathia $204.47 J. Smoltz $200.58
Not surprisingly, Johan Santana leads the pack. What does a price of $266 equate to? Let’s take Santana’s 2006 season when he had 19 wins, 245 strikeouts, 220 innings, 0.997 WHIP, 2.77 ERA and 686 recorded outs. Doing the math translates to a price of $266! The market expects Johan's 2007 season to be bang on equal to his 2006—his stats so far indicate that this is fair.
In contrast, here are the 10 cheapest pitchers:
Stock Price K. Davies $43.15 D. Marte $51.99 B. Medders $52.18 S. Downs $52.73 J. Benoit $53.09 A. Embree $53.73 W. Ohman $54.27 R. Tejeda $54.30 J. Rauch $54.70 J. Grabow $54.97
Is Kyle Davies worth more than $43? As a marker his rookie season of seven wins, 4.93 ERA and 1.68 WHIP works out at $32; his 2006 numbers (ERA 8.38) would require a visit to Chapter 11—the market expects a significant improvement!
Here are the most and least expensive hitters according to PROTRADE:
TOP 10 BOTTOM 10 Stock Price Stock Price A. Rodriguez $365.58 B. Williams $71.48 J. Reyes $361.63 C. Guzman $74.81 D. Ortiz $340.82 B. Schneider $81.75 D. Wright $333.90 A. Callaspo $81.90 D. Jeter $332.09 K. Matsui $82.95 C. Beltran $330.55 N. Logan $87.31 A. Pujols $327.36 D. Erstad $87.63 G. Sizemore $324.22 M. Diaz $88.01 T. Hafner $320.87 C. Sullivan $88.01 D. Lee $319.41 C. Kotchman $89.48
I suppose the list won't surprise too many. It's interesting to note how close in price A-Rod is to Jose Reyes. This brings out one of the foibles of the fantasy scoring system we alluded to earlier—namely the relative value of different offensive events. Reyes' price is punchy not only because he is a good hitter (he is, but nowhere near Planet A-Rod) but also because he'll commit larceny on the basepaths.
A couple of benchmarks for you: a bench player batting .250, 23 runs/RBIs, five home runs wouldn't get much change from $30; a platoon specialist hitting .270, with 50 runs/RBIs and 15 home runs is priced at $94; while a .300 hitter, with 120 runs/RBIs, 50 home runs and 20 steals is worth $310. One thing is clear: given that A-Rod doesn't steal much PROTRADE thinks he will have a season for the ages.
As well as player markets, PROTRADE has also launched team markets. The concept is the same but based on how well buyers and sellers think a particular team will perform.
Here is the scoring system:
Price Criteria $1 for each regular season win $10 for qualifying for the playoffs $4 for each playoff win $10 for winning the league division series $20 for winning the pennant $30 for winning the World Series
And here is price for each team (as of Saturday April 21):
Stock Price Washington Nationals $60.32 Kansas City Royals $64.89 Seattle Mariners $76.29 Baltimore Orioles $77.79 Tampa Bay Devil Rays $78.95 Pittsburgh Pirates $79.13 Cincinnati Reds $80.19 San Francisco Giants $80.21 Florida Marlins $82.21 Colorado Rockies $83.54 Texas Rangers $84.82 Toronto Blue Jays $91.13 Chicago White Sox $91.64 Houston Astros $92.51 Philadelphia Phillies $96.55 Chicago Cubs $97.25 Arizona Diamondbacks $99.47 St. Louis Cardinals $100.64 Detroit Tigers $101.39 Milwaukee Brewers $101.40 Los Angeles Angels $101.80 San Diego Padres $102.57 Atlanta Braves $104.06 Oakland Athletics $105.13 Cleveland Indians $107.26 Los Angeles Dodgers $110.11 Minnesota Twins $113.21 New York Mets $131.98 Boston Red Sox $132.93 New York Yankees $137.19
It isn't too difficult to develop a rudimentary valuation model for teams. Using a logistical regression it is possible to work out the probability of an X-win team making the playoffs. For instance, a 90-win team will make the postseason 57% of the time. Here are the playoff odds across a range of win levels:
Wins Playoff Odds 84 7% 85 11% 86 17% 87 25% 88 34% 89 46% 90 57% 91 68% 92 77% 93 85% 94 90% 95 93% 96 96%
Using this in conjunction with the points system allows us to piece together what different price ranges imply about a team's playoff chances.
- Less than $85: Realistically any team in this bucket doesn't have a great shot of playing October ball. The price is equal to the number of wins they’ll get. Teams closer to the $85 mark will have a small chance of going to the postseason priced in.
- $85-$95: Fringe playoff candidates, perhaps a 5-10% chance but don’t bet your mortgage
- $95-$110: Strong playoff candidates—20-30% chance of making it to the postseason though likely to get knocked out in the first round
- $110-$140: Very likely to make the postseason and a good chance to progress to the League Championship and, possibly, the World Series
- Over $140: Racing certainties for a pennant and World Series appearance.
Any team that makes the World Series is going to earn a minimum of $150 (assuming a 92-win season). The highest price on our chart is $138 for the Yankees, which means the market thinks they have a good shot at making the World Series (the price should reflect the risk that they don't).
The right way to dissect market expectations is to pull together some scenarios as I have done below for the Atlanta Braves.
Braves' Team Price
The Braves are priced at $103, which means that there is a strong expectation that they make the postseason (consult guidance above). Let’s construct five scenarios:
- Win 85 games and don’t make playoffs
- Win 92 games and make playoffs but get knocked out in NLDS
- Win 92 games and get knocked out in NLCS
- Win 92 games and get knocked out in World Series
- Win 92 games and win World Series
There are a couple of things to note. First is that if you’re a borderline playoff team the number of games won isn’t that important as postseason performance drives a lot of extra value. Second, the expectation of whether you qualify for the postseason can cause a massive inflexion in price around the $90-$95 mark (as this is how many wins a team typically needs to have a decent shot of making the playoffs).
Using the scoring system we can work out the worth of each scenario and then assign probabilities. The THT division standings gave the Braves a 24% chance of making the postseason. Given their hot start this may have crept up a little—let’s call it 30%. We'll divide the balance (70%) among the other four options. Here are the scenarios with price and odds:
Scenario Price Odds 1 Win 85, don't make postseason 85 70% 2 Win 92, knocked out in NLDS 110 12% 3 Win 92, knocked out in NLCS 132 8% 4 Win 92, knocked out in WS 168 6% 5 Win 92, win WS 206 4%
All this gives a probability weighted price of $102, which values the Braves fairly (current price is $103). By flexing the scenarios it's possible to paint a picture of some of the upside and downside risks to the price. For instance, if the Braves went on a losing streak and their probability of making the postseason dropped to 10% then the price would fall to around $85-$90.
Other Thoughts on Team Markets
Here are some other observations from the team data:
- Comparing the AL to the NL it is clear that there is more uncertainty about the the outcome of the junior circuit—we can see this because a lot of NL teams hover around the $95 price mark
- Expect a Red Sox and Yankees ALCS—the Yankees are slight favorites
- A subway series is a strong possibility as the Mets are strongly fancied to win the NLCS
- Elsewhere in the NL, the Dodgers are early favorites to take the West, while the Brewers and Cardinals are in a two way tussle for the Central, and Atlanta are adjudged likely Wild Card champs
- The AL looks more overpriced as it contains five of the seven most expensive teams teams—the Indians, Twins, Tigers, Angels and Athletics are all expected to join the Red Sox and Yankees in the playoffs. If you believe that the wild card will go to the East only two of those five teams will make it—I’m a seller.
PROTRADE is another example of markets in action, albeit tailored to more of a fantasy audience.
As the site expands and becomes more popular I’m sure you’ll see new markets pop up. Personally, I’d like to see win-based markets, and also markets to evaluate the likely outcome of trades. I’d also like to see career markets too to try to evaluate prospects. And In the future it'd be cool to trade derivatives, for instance, AL third basemen, or a subset of uber-prospects.
PROTRADE is a professional, well financed outfit with some talented executives—over time I expect them to take sports prediction markets to the next level. I’ll keep checking in with some of the goings on as the year progresses. Stay tuned.
References and Resources
Needless to say this column wouldn't have been possible without the PROTRADE website—presumably if you've gotten this far you realize that.
John is an unashamed glory supporter having followed the Atlanta Braves since 1991. He blogs the Braves at Chop-n-Change. He welcomes comments, criticisms and suggestions via e-mail
<< Return to Article