The column labeled "Avg Juice" takes the total amount of money I would have won on each team divided by the amount risked. The next two columns are each team's third order wins and losses. The penultimate column is third order winning percentage and the last column converts that winning percentage into Vegas Odds. In the last column, green means I have been consistently getting better odds than the third order record would suggest, while red is the opposite. (UPDATE: Per request, I put my +/- units in the last column) I only looked at teams I've bet on more than 5 times, and I didn't control for statistical significance in the color coding.
What I find most interesting is that three of the four teams that have taken most of my money are the teams that I haven't been getting better than third order juice on my wagers (with statistical significance issues lingering, particularly with Washington). My initial inclination was that these teams were underperforming, and that I would have at least been getting value recently as their juice should have increased. To test this, I looked at the time series of Washington's juice over the season. The trend was not significant. Same with Oakland.
I'm not entirely sure what this means, other than I have likely been betting on Washington (n=34) and Oakland (n=25) too much, when they don't actually have value. Of course, my other bets are not necessarily +EV because I didn't look at pitching matchups, etc. Particularly because of pitching matchups, I would expect a positive bias to these results, since in Contrarianville, we often bet on the "wrong" side of pitching mismatches. Still, overall, I'm happy with these results.
(UPDATE2: Going off ilike#s comment, here are two graphs comparing the "value" and my unit profit. They show a weak positive correlation.)
10 comments:
This is great stuff, well done. Can you add a column with your Net $ for each team (scrolling is hard). It would interesting to see the profits for each team compared to the ave-juice vs. O3ave-juice. Interesting nonetheless.
Its odd the correlation got worse when the Nats were omitted. Is that really correct?
I think so. Outliers at tails of distributions can really affect regression results, particularly with SSS. That said, there is no reason not to include the Nats either.
By definition, shouldn't an outlier decrease correlation?
"What I find most interesting is that three of the four teams that have taken most of my money are the teams that I haven't been getting better than third order juice on my wagers"
Of course it works out like this. The Nationals played poorly in April, and you lost money on them. Since they had played poorly, you kept betting on them, and they played poorly again in May. It's not like there isn't a correlation between actual W/L results and third order record, so this analysis is probably going to indicate that a team that's played poorly has not had "value". Which is not necessarily the case, and does not prove anything. And it certainly doesn't prove this:
"...I have likely been betting on Washington (n=34) and Oakland (n=25) too much, when they don't actually have value."
This is one step away from saying that you can tell they didn't have value because they lost. Third-order record =/= what the "true" odds were on those games.
It's not like there isn't a correlation between actual W/L results and third order record, so this analysis is probably going to indicate that a team that's played poorly has not had "value".
Well, yeah, except that Matchbook gamblers have them pegged as a .430 team in the 35 games I've wagered (with no trend to the juice) and PCT3 had them pegged as a .430 team through 49 games (when pitching matchups and HFA are smoothed out over the sample). Is PCT3 not the expected number of wins and losses given their base runs and SoS? If expected winning percentage = odds received, how is that not a sign of no value? What am I missing here?
Third-order record =/= what the "true" odds were on those games.
I obviously agree on an individual game (again: pitchers, HFA, etc.), but I fail to see why this is true when looking at moderate sized sample.
"If expected winning percentage = odds received, how is that not a sign of no value? What am I missing here?"
Because the expected winning % measures what their record should be based on what they actually did not what they theoretically should have done based on their talent level, if that makes sense.
Because the expected winning % measures what their record should be based on what they actually did not what they theoretically should have done based on their talent level, if that makes sense.
Ok, that makes sense. So if I had used these regression equations instead, I'd have a better metric?
Also, I'd like to point out that this conversation is moot for everyone except Washington and probably Oakland. Everyone else is too SSS to make any sort of statistical conclusion about value I got on the team because of all of the other variables, even if the intent of the post was in the right place. But since I/we bet on WAS and OAK a lot, I'd like to get this hashed out.
"Is PCT3 not the expected number of wins and losses given their base runs and SoS?"
No, that's PCT2. PCT3 is against a neutral schedule. If you're going to use this you have to use PCT2.
"I obviously agree on an individual game (again: pitchers, HFA, etc.), but I fail to see why this is true when looking at moderate sized sample."
Because it's still not a big enough sample. You're saying the Nats have not had value because they have played poorly over 49 games. That's not a conclusion you can draw, the theoretical margins are just too small. The Nats' true W% over those games could have EASILY been .440, and then suddenly there is value. Again, it's one step away from saying they haven't had value because they've lost a bunch of games.
"Because the expected winning % measures what their record should be based on what they actually did not what they theoretically should have done based on their talent level."
This.
Post a Comment