Hoop Pointers: Valuing the Trade
March 21, 1998
In my last essay, I evaluated the value and importance of the opening draft. I did this by examining the top 50 teams. My general conclusion was that, at least among these lofty teams, there was no discernable relationship between a "good draft" and a "good finish", as the drafts of the top 50 teams were all over the lot, whether valued in dollars or points. This is not a surprising result, since the number of trades allotted to each teams permits a roster to be completely turned over four times during a season. Also, since a good deal of trading occurs very early in the season, the impact of the draft is usually short lived.
If the draft isn't all that important, then it must be trades that make the difference. Is there a meaningful way to benchmark the value of a "good trade"?
Here's how I tackled the question. Again, considering only the top 50 teams, I compared both the points and the roster values for the initially drafted roster to the current roster. Then, I divided the differences by the corresponding number of trades for each team. Let's look at the two methods and the results.
Using prices as of March 5th, the average roster value for the top 50 teams was $133 million. In my last essay, I reported that the corresponding average value of these teams' drafts (using March 5 prices) was about $58 million. So, trading seems to have increased roster values by an average of $75 million. The average team had used 49 trades as of this date. Dividing the $75 million gain by 49 trades produces an average gain per trade of just over $1.5 million. Looking at this calculation team by team, I find that the best of these teams averaged about $2 million per trade, while only $1 million per trade was achieved at the other extreme. Interestingly, there was no correlation (within this group of 50) between the average value per trade and accumulated Smallworld points. In fact, it appears that some of the best traders were the worst drafters, so one skill seems to have compensated for any shortfall in the other.
Of course, while generating roster value is a critical means of picking up better players, the real object of the game is to generate Smallworld Points (SWP). For this measurement, I used projected values for SWPs, rather than current values. I projected the season ending tally of SWPs for each team by assuming that each current player would continue to produce for the rest of the season at his average SWP/game pace. For comparison, I did the same projection using the initially drafted rosters. The difference between the two projections is the projected SWP gain attributable to trading activity.
For these 50 teams, the average projected SWP total is just under 37,000. For the draft rosters, the corresponding projection is 24,000. The difference, which is the extra SWPs generated because of trading, is 13,000. Again, dividing by an average of 49 trades, the average SWP value of a trade works out to be 265 SWP. Averages can be a bit difficult to translate into "per game" equivalents, since a trade done at the beginning of the year would only need to produce an extra 4 SWP/game to meet this standard, while a trade done at mid year would need twice that improvement. I'm sure the average trade date for these teams is well before mid-season, maybe closer to the 25% point. It's probably reasonable to assume, then, that the average trade produced about a 6 SWP/game pick-up.
Are there any lessons here? First, to be a worldwide contender, it helps to keep these benchmarks in mind when planning your trading strategies. I've noticed that a lot of managers do a flurry of trading early in the season, apparently just to gain exposure to an extra game or two (selling players with one or two days off, and buying players with immediate games scheduled.) To be worth the average of 265 SWP/trade, you would need to gain an additional six games for a player who averaged 44 SWP/game, and a greater game increase is need for players with more modest SWP/game expectations. So that strategy doesn't seem to be a winner, does it?
As I've written before, early trading should be for the purpose of increasing value. I've used a rule of thumb that a trade should only be done if the expected gain is at least $1.5 million, and the averages of the top 50 teams seem to confirm this as a good "stretch" objective. If you can manage to this average, the additional SWPs should come, especially if you build your roster value quickly enough.
Obviously, it's too late to do much about this for the remainder of the hoops season. And the Smallworld baseball format, while similar in broad respects, will use a different roster size, and a different repricing frequency, so these averages might not be appropriate benchmarks for baseball. But if you're going to improve, you need to learn from your mistakes. And perhaps, by reflecting on your trading style in light of these "best practice" averages, you'll be better equipped to improve your standing the next time around.
Return to Guru's Hoop Pointers index
Hoop Pointers is written by Dave Hall (a.k.a. the Guru), an avid fantasy sports player. He is not an employee Smallworld, and any opinions expressed are solely his own. Questions or comments are welcome, and should be emailed toGuru<firstname.lastname@example.org>.