Poster Michelangelo wants to bet Vitterd $1K on baseball for the rest of the season. Vit won't take the bet. Why?

[Ricboff]

Reversion to the mean, also called regression to the mean, is the statistical phenomenon stating that the greater the deviation of a random variate from its mean, the greater the probability that the next measured variate will deviate less far. In other words, an extreme event is likely to be followed by a less extreme event.
Although this phenomenon appears to violate the definition of independent events, it simply reflects the fact that the probability density function

of any random variable

, by definition, is nonnegative over every interval and integrates to one over the interval

. Thus, as you move away from the mean, the proportion of the distribution that lies closer to the mean than you do increases continuously. Formally,

int_(mu-j)^(mu+j)P(x)dx " style="border: 0px;">

<tbody>
</tbody>

for

j>0" style="border: 0px; vertical-align: middle;">.
The Season 1 episode "Sniper Zero" (2005) of the television crime drama NUMB3RS mentions regression to the mean.


...........

 

[vitterd]

I can understand thinking you are right when it is you and Zit going back and forth since you guys have that tension thing going. But when 3 or 4 people tell you that your logic may be flawed, why not just re-evaluate your thought process? I don't understand why this is so hard for some people.

There is nothing to reevaluate. If he hits 55 percent over the second half and I hit 54 percent...he loses a ton for the year and I win a ton....but I lose the bet. Then he's the better capper for the baseball season? I pay him a dime while I win 10k on the year and he loses 50k

what is so hard to understand about that?

 

[PatsFan1283]

Regression to the mean doesn't matter in this case. All that matters is theoretical win-rate of the two combatants.

Of course he is likely to regress to the mean, doesn't mean he isn't +EV if his theoretical win-rate is higher than the other person.

 

[Ricboff]

k. I think that's his angle. Or perhaps i'm wrong

 

[MantisToboggan]

You think I will continue this rate for the rest of the year? I do not. There is no reason to think I will hit this high rate over a full season. There is no reason to think he will finish this low. I will come back to the pack and he will move up. I will destroy him for the overall season....and will lose 2nd half bet in all probability. Im not paying him a dime for that.

No, I have no clue what rate you will actually hit at for the rest of the season. If you're both 55% guys, as you claim, then odds are you will each hit at a 55% clip the rest of the year. Let me put it to you like this... Let's say we are making 100 plays and are both 50% cappers. The first 10 I go 10-0 while you go 0-10. On the next 90 plays, it's still most likely that we each go 45-45.
If you think you're a better baseball bettor than he is, you should take the bet regardless of your first half or his first half.

 

[vitterd]

No, I have no clue what rate you will actually hit at for the rest of the season. If you're both 55% guys, as you claim, then odds are you will each hit at a 55% clip the rest of the year. Let me put it to you like this... Let's say we are making 100 plays and are both 50% cappers. The first 10 I go 10-0 while you go 0-10. On the next 90 plays, it's still most likely that we each go 45-45.
If you think you're a better baseball bettor than he is, you should take the bet regardless of your first half or his first half.

i understand what you're trying to say but you aren't listening to what I'm saying. If he outperforms me for the second.....by even a little....he wins the bet. Essentially he could lose 59k in bets for the year and I could win 12k and he wins the contest because it doesn't account for a full year. I'm simply not doing that.

 

[MantisToboggan]

i understand what you're trying to say but you aren't listening to what I'm saying. If he outperforms me for the second.....by even a little....he wins the bet. Essentially he could lose 59k in bets for the year and I could win 12k and he wins the contest because it doesn't account for a full year. I'm simply not doing that.

That's fine, but if you think you're a better baseball bettor you're wasting an opportunity to make money.

 

[vitterd]

That's fine, but if you think you're a better baseball bettor you're wasting an opportunity to make money.

I am a better baseball capper than him....but not as much as the first half shows. Like I said, he could beat me in 2nd half while I smash him overall for the season and that simply isn't fair. I'm not paying him a thousand and losing the contest while I prove to have the better year by far. It would be silly.

 

[festeringZit]

This shit is so funny. He still has absolutely no clue.

 

[vitterd]

This shit is so funny. He still has absolutely no clue.

Youre not grasping this at all. I'm not paying a guy a dime that I beat in baseball by 60 dimes because he outperforms me by a little in the 2nd half. No sane person would do that.

 

[Ricboff]

if Vit is hitting 75% ytd. And historically , over the last 15 years is a 55% baseball winner from start of season to end of season.

then, well this COULD be an outlier year.... however....probability is not on his side. Regression to the mean.


If MIch is hitting 35% ytd. And historically , over the last 15 yrs is a 55% baseball winner from start of season to end of season

then, well this could be an outlier year........however........regression to the mean says the 2nd half 'should' be better.


and yes their next baseball bets know nothing of this. I get that.


thoughts?

 

[PatsFan1283]

Assuming the betting environment is the same as previous betting environments, the better capper is the fav Ric. Regardless of recent short-term results that the two have had.

 

[vitterd]

if Vit is hitting 75% ytd. And historically , over the last 15 years is a 55% baseball winner from start of season to end of season.

then, well this COULD be an outlier year.... however....probability is not on his side. Regression to the mean.


If MIch is hitting 35% ytd. And historically , over the last 15 yrs is a 55% baseball winner from start of season to end of season

then, well this could be an outlier year........however........regression to the mean says the 2nd half 'should' be better.


and yes their next baseball bets know nothing of this. I get that.


thoughts?

Of course. Somebody else gets it.

Thats not the only reason for not taking the bet. He could outperform me by 1 dollar over the 2nd half. So, I still outperform him by a large margin over the full season but then owe him a thousand dollars and lose the contest.

It would simply be an idiotic bet for me just based on that.

 

[festeringZit]

You have no comprehension of basic gambling concepts like expected value. None. But keep posting here because this shit is really funny.

 

[vitterd]

You have no comprehension of basic gambling concepts like expected value. None. But keep posting here because this shit is really funny.

You don't grasp anything. I'm not paying a guy I mash for an entire season because he outperforms me by a dollar in the 2nd half.

vit ends season at plus 13k

mich ends season at -59k

vit owes mich 1,000 and he wins and is the better baseball handicapper. Yeah....ok face)(*^%

 

[Ricboff]

Assuming the betting environment is the same as previous betting environments, the better capper is the fav Ric. Regardless of recent short-term results that the two have had.

okay, say Vit ,as he claims is the better capper. Say 55% to 53%. So in post #91-- leave Vit's career baseball ytd winning % at 55, but change Mich's to 53%



given ytd numbers (post 91), still a fair bet (given historical data)?

 

[MantisToboggan]

okay, say Vit ,as he claims is the better capper. Say 55% to 53%. So in post #91-- leave Vit's career baseball ytd winning % at 55, but change Mich's to 53%

given ytd numbers (post 91), still a fair bet?

Yes, regression just means they are likely to go back to their averages. Vit hits 55% and Mich hits 53% the rest of the way.

 

[PatsFan1283]

okay, say Vit ,as he claims is the better capper. Say 55% to 53%. So in post #91-- leave Vit's career baseball ytd winning % at 55, but change Mich's to 53%



given ytd numbers (post 91), still a fair bet (given historical data)?

Yeah, using % isn't the right way to look at it with baseball because they aren't spread bets.

The better way to look at it is ROI on their baseball bets. Whoever is better at generating a positive ROI would be the theoretical fav in the bet. Regardless of past results.

Now baseball and handicapping and the markets etc are always evolving and new sample sizes are constantly needed to measure success, so it can be tough to truly say how good someone is at these endeavors.

This is a big reason John Bogle recommends index investing, because even if someone can beat the market for a period of time, you likely won't be able to find them in time and just going off past results probably isn't enough to say they can produce a superior return. It could've just been variance.

 

[vitterd]

Yes, regression just means they are likely to go back to their averages. Vit hits 55% and Mich hits 53% the rest of the way.

Lets say mich hits 55 and I hit 53. He wins 2nd half bet while I win by large margin for the season. You really can't see how that would be an incredibly stupid bet for me.

Even throwing out regression.....it would be a stupid bet.

 

[MantisToboggan]

Lets say mich hits 55 and I hit 53. He wins 2nd half bet while I win by large margin for the season. You really can't see how that would be an incredibly stupid bet for me.

Even throwing out regression.....it would be a stupid bet.

No, I don't think it's a stupid bet if you are the better baseball bettor. I think it's stupid not to take the bet.