Danshan on Kelly again

[AsianmanSports]

Thats fair.

[tsty]

It is not fair

It is very far off base

[AsianmanSports]

Care to elaborate?

[tsty]

I agree with you and Tommie that full Kelly is the way to go. What do you think about using the lower end of a 95% confidence interval to use as the estimate for Full Kelly bets?

That is perfectly fine and I would even be willing to go lower than that

Most of this stuff is easy to simulate anyways and being wrong up to a certain point wont even affect you in the long run

Its retarded to assume overbetting 2% or 4% or x% will be the difference between a 100k year and 50k year after thousands of bets

You do not need to be 100% accurate for kelly to be the optimal choice let alone the superior choice compared to flat betting

Flat betting is just straight up awful

[danshan11]

Kelly is impossible to backtest unless you use a fixed margin for all backtesting and if you do use a fixed margin on every bet, you are now flat betting!

[danshan11]

That is perfectly fine and I would even be willing to go lower than that

Most of this stuff is easy to simulate anyways and being wrong up to a certain point wont even affect you in the long run

Its retarded to assume overbetting 2% or 4% or x% will be the difference between a 100k year and 50k year after thousands of bets

You do not need to be 100% accurate for kelly to be the optimal choice let alone the superior choice compared to flat betting

Flat betting is just straight up awful

why is flat betting awful?

[AsianmanSports]

When betting aprox 50 - 50 odds flat betting is perfectly fine.

[Gaze73]

If none of y'all can get even 5% roi on closing lines then y'all are squares, nuff said.

[danshan11]

If none of y'all can get even 5% roi on closing lines then y'all are squares, nuff said.

can you first give me an example of what you mean by getting a 5% roi on closing lines?

[AsianmanSports]

If none of y'all can get even 5% roi on closing lines then y'all are squares, nuff said.

lol

[Gaze73]

can you first give me an example of what you mean by getting a 5% roi on closing lines?

Hit 55% at -110 one hour before the game.

[danshan11]

Hit 55% at -110 one hour before the game.

Gaze if you only understood what skill level it would require to do that long term. and 2nd even if you did that can you please show me your math on how that equates to a 5% ROI?

[AsianmanSports]

Hit 55% at -110 one hour before the game.

55% Win at -110 = 5% Roi? Wut?

[Gaze73]

Y'all don't even know basic math. -110 = 1.91 so 0.55*1.91 = 1.05 = 5% roi

[danshan11]

Y'all don't even know basic math. -110 = 1.91 so 0.55*1.91 = 1.05 = 5% roi

thanks Gaze LOL

[AsianmanSports]

lol, f*** me

[danshan11]

lol, f*** me

I did not understand what he meant, I thought he had some master math, LOL and he is right I guess I dont know basic math, wait 1+1 is???? ughh 2???


he still is in la la land if he thinks its a piece of cake to win 55% ATS, that part is fantasy

[TommieGunshot]

that is because you think you know your edge, herein lies the problem, if you say man I got the Blueballs -190 and the book has them at +120, that is a huge bet with kelly when the Blueballs are probably really worth +120, you are thinking you have this huge edge and betting accordingly, this is flawed UNLESS you are a master capper who has a better line modeled than the entire betting consensus. how many of these master cappers are laying around?

When the market is offering a price on a team of +120, the implied odds are that they have a less than 45% chance of winning (most likely around 43-44%). Betting them to win at +140 would be a roughly 5% edge, based on those implied odds. (I would act conservatively, understanding the market may not be true odds and consider the edge to be quite a bit less than that). Betting them at +150 would be an even bigger edge. (Betting them at -190 would be about a negative 60% edge). Not betting more when the edge becomes larger is simply minimizing profits. Obviously people who don't understand the math behind Kelly Criterion, don't know how to estimate probabilities, or have other things going on will have to minimize their profits.

you say I am not here to learn, then teach me something. What am i missing? explain it to me. now if you are talking about line shopping and using the current line to base your kelly%, you probably can have some success since we know the current line is the most efficient. the problem is most kelly bettors are not using the line to base their bet size on, they use their historic edge, their model edge and whatever other crazy ideas they can come up with!
Vote NO on Kelly!

All of this is based entirely upon projecting your own silly ideas onto everyone else, even when they are completely wrong. Pretty much everyone I've met who profits from betting sports uses market prices as a major factor (often the only factor) in estimating probabilities. They then have plenty of outs to always get a good price, if not the best price available.

[danshan11]

can you list the silly ideas and the correct answer, that would be really helpful.

if you use the closing line to estimate probability you cant bet cause the line is closed. so you use the current line to estimate probability and if it closes at that point kelly worked perfect if not well you know!

when you say minimize profits you mean having a bad model number and that is why you are so far off. if you are modeling +120 and the line is +170, you need to stop worrying about Kelly and start worrying about your model.

really interested for you to list my silly ideas and the correct answers to those silly ideas, that would be really helpful!

[TommieGunshot]

really interested for you to list my silly ideas and the correct answers to those silly ideas, that would be really helpful!

You come up with an example of where you agree Kelly Criterion would be the best way to maximize profits ("now if you are talking about line shopping and using the current line to base your kelly%, you probably can have some success since we know the current line is the most efficient"); but because you are either unwilling or unable to apply anything from that example (too lazy? too hardheaded? unwilling to learn?) you dismiss the entire formula despite being supported with rigorous math.

[danshan11]

what rigorous math supports using kelly on an unknown probability?

and you said silly ideas, what else is silly? Please list them one by one if you would.

[TommieGunshot]

what rigorous math supports using kelly on an unknown probability?
and you said silly ideas, what else is silly? Please list them one by one if you would.

It is used when the probability is known. The rigorous math is easy to find: https://en.wikipedia.org/wiki/Kelly_criterion#Proof. And you said yourself that when using market prices to assign probabilities, Kelly Criterion can be used to maximize profits. (Again: "now if you are talking about line shopping and using the current line to base your kelly%, you probably can have some success since we know the current line is the most efficient"). It is silly to then turn around and say that Kelly Criterion cannot be used to maximize profits.

But it's obvious you're only here to argue and not to learn.

[tsty]

even a penetrating simple steam chase example goes over your head

also stop using "unknown probability" since using your terminology then even a coin toss is an unknown probability

just so penetrating stupid taking things to retarded extremes

you are 50 years old from a finance background too? what in the actual penetrate

[danshan11]

even a penetrating simple steam chase example goes over your head

also stop using "unknown probability" since using your terminology then even a coin toss is an unknown probability

just so penetrating stupid taking things to retarded extremes

you are 50 years old from a finance background too? what in the actual penetrate

a coin flip is an unknown probability?

[danshan11]

It is used when the probability is known. The rigorous math is easy to find: https://en.wikipedia.org/wiki/Kelly_criterion#Proof. And you said yourself that when using market prices to assign probabilities, Kelly Criterion can be used to maximize profits. (Again: "now if you are talking about line shopping and using the current line to base your kelly%, you probably can have some success since we know the current line is the most efficient"). It is silly to then turn around and say that Kelly Criterion cannot be used to maximize profits.

But it's obvious you're only here to argue and not to learn.

I dont believe that in sports betting we know the true probability of ANY game

[tsty]

a coin flip is an unknown probability?

Yes it is just an estimation...

If you dont even understand this lol

[tsty]

I dont believe that in sports betting we know the true probability of ANY game

This is the same thing for everything in life

Including flipping a coin

[danshan11]

let me understand clearly are you saying we dont know the probability of a coin flip barring nuclear holocaust?

[tsty]

let me understand clearly are you saying we dont know the probability of a coin flip barring nuclear holocaust?

Use the same logic you are using for sportsbetting

It can also apply to flipping a coin

I am just trying to use simple examples to show why you are wrong but you seem to just ignore everything

Anyways i am done trying to help you

I am not sure how you got to the age of 50 but god help us all

[danshan11]

I hope I am not ignoring anything, I try and be responsive and inquisitive at anything I can get! Sometimes the things are just too complicated for me to get, sorry!

[TommieGunshot]

I dont believe that in sports betting we know the true probability of ANY game

We can use implied probabilities to come up with estimates and ranges that can give us a very good guide to what the edge is on any given bet. To not bet bigger when the edge is bigger means minimizing profits.

Now we all fully expect you to come up with some comment saying why you disagree. It would be interesting for someone to post probabilities on that for us to bet on.

[danshan11]

We can use implied probabilities to come up with estimates and ranges that can give us a very good guide to what the edge is on any given bet. To not bet bigger when the edge is bigger means minimizing profits.

Now we all fully expect you to come up with some comment saying why you disagree. It would be interesting for someone to post probabilities on that for us to bet on.

the BOLD sounds like a bunch of maybes and that may be cool and work, I dont have evidence to the contrary but I just think with maybes you are probably better off flat betting.

[pretentiousGuy]

If none of y'all can get even 5% roi on closing lines then y'all are squares, nuff said.

I'm sorry, but how long have you been doing this for exactly? At least a few years judging by your post history, but your bet size is in the 10s of dollars?

Why are you so confident in your betting ability?

[tsty]

I'm sorry, but how long have you been doing this for exactly? At least a few years judging by your post history, but your bet size is in the 10s of dollars?

Why are you so confident in your betting ability?

Lol say it aint so

I knew gaze is bad but cmon

Also it is impossible to beat the big market closers for that much

It is just the nature of the beast

If you could then you should easily be making millions year in year out

[Believe_EMT]

and just so everyone knows it and hears it, kelly sucks, use kelly to figure out your grocery bill or something but NEVER in sports betting, where you never truly know your edge.

holy fukk man! i have been saying this same thing for years and everyone poops on it!

the entire thing is built on the faulty assumption that any of us can accurately forecast our edge every game.