Jameis Winston alleged point-shaving

[GVC]

I think you can take that theory a step further, and talk about the distribution of points scored, especially if player props lines are at or above players' scoring averages. I'd guess it's more likely a player scores 10+ points over his average than 10+ points under his average (especially for players with low averages). Here, the variance on scoring conditional on scoring above your average is greater than the variance on scoring conditional being below your average. Further, the mean conditional on being below your average is probably closer to your average than the mean conditional on scoring above your average. This all implies (well, mostly the second point) that players are more likely to score below their average than above their average (while also being more likely to have outliers above their average than below...in part because you can't score fewer than 0 points).

This can all be tested empirically, of course. Maybe I'll do this for Jazz players last season.

 

[QuinSnydersHair]

I think you can take that theory a step further, and talk about the distribution of points scored, especially if player props lines are at or above players' scoring averages. I'd guess it's more likely a player scores 10+ points over his average than 10+ points under his average (especially for players with low averages). Here, the variance on scoring conditional on scoring above your average is greater than the variance on scoring conditional being below your average. Further, the mean conditional on being below your average is probably closer to your average than the mean conditional on scoring above your average. This all implies (well, mostly the second point) that players are more likely to score below their average than above their average (while also being more likely to have outliers above their average than below...in part because you can't score fewer than 0 points).

This can all be tested empirically, of course. Maybe I'll do this for Jazz players last season.

I completely agree.

FWIW, I looked at Dirk. He scored 21.7 ppg last year. If oddsmakers set his number at 21.5 on average, he went under it 43 times and over it 39. Not exactly great, especially when you consider in the vig. But that's why the key is to do the under on all players listed.

 

[GVC]

I completely agree.

FWIW, I looked at Dirk. He scored 21.7 ppg last year. If oddsmakers set his number at 21.5 on average, he went under it 43 times and over it 39. Not exactly great, especially when you consider in the vig. But that's why the key is to do the under on all players listed.

What percentage of bets do you generally have to win to break even?

If the number is based on predicted season average, betting on multiple players won't give you any advantage over betting on a single player.

 

[GVC]

Since I don't bet, I have another basic question: Are betting lines generally set to get even money on both sides of a line? That is, are they just making money on the vig (or do they take on some risk)?

 

[QuinSnydersHair]

What percentage of bets do you generally have to win to break even?

If the number is based on predicted season average, betting on multiple players won't give you any advantage over betting on a single player.

To break even, between 55-60%.

It will benefit me in the sense that I can't predict which 1-4 players will score well above their averages. If I just did Dirk or just did Ellis, there's a solid chance they go over. There is not however a solid chance that 3 of 4 or 4 of 4 go over.

 

[QuinSnydersHair]

So rather than just pick a random player on a given night, I do all of them and stick consistently with that method.

 

[GVC]

Shouldn't make any difference if the line is based on predicted season scoring average. That is, if you can bet on the same players every game, and (again) the line is based on predicted season scoring average, your expected winning percentage will be the same whether you bet on 1 or 30 randomly selected players. Your winnings will just be less volatile.

 

[GVC]

To break even, between 55-60%.

Then you'll need quite a bit of dumb money pushing the line up to make bets on the under worthwhile, based on the distribution of points scored by the young Jazz players last season.

 

[QuinSnydersHair]

Will respond in a bit. Screaming baby on a car ride home. Lol.

 

[TroutBum]

Well, you guys took a pretty fun subject like sports betting and turned it into a ****ing math equation. I hate both of you.

 

[VINYLONE]

That's nothing. I had about 200-250 last week.

Up 4,500 the last three weeks!

I actually have a very sensible NBA player props theory I'm trying out this year. Did it for the first time last night and went 4-0.

Judas priest. I don't have a reference at all for wagering, so is 200+ wagers a week a lot or a little? Are they are all seperate wagers? Where do you find the time to do the research? Also, since I highly doubt you started betting 3 weeks ago, do you keep a spreadsheet or table tracking your winnings? You say you're up 4.5k in the last 3 weeks? Is that from 3 weeks ago or from the inception of your wagering? What's the biggest amount you've lost?