Well I'm admittedly probably the last reg on here that should be giving Kelly advice, but yea...find +EV bets and bet the appropriate amount and you should be good.

How do you know your bets are +EV though?

^That's a question I don't want to get into because it's not where my shortcoming is. It's on the issue of staking that I'm uncertain.

Interesting. I think I might do this then.
Do you think a case could be made for using full Kelly's if the bettor happens to specialise in a sport that only has about 30-40 events weekly (worldwide)?

It is Kelly Criterion. Not Kelly's Criterion.

Have you spent anytime looking at the subject? It sounds like you don't really even understand what questions to ask.

wrt your question. No. That doesn't make any sense. The overall number of events is irrelevant.

It is actually amazing you still don't get this after years.

This may be helpful:

http://talkingnext.io/betsize-and-simulations

That link would be the opposite of helpful.

Kelly is path independent. Sequencing is irrelevant.

There is a mountain of stupid written about Kelly by people who're too stupid or too lazy to stumble across the right answer. Or more than likely don't work with a bunch of people smarter than them to help un**** their stupidity.

Some questions:

#1 - When websites track whether a team has gone ATS or not, which bookie do they use as the final authority? Is it the very best line they could find, having considered all bookies?

#2 - When websites track whether a team has gone ATS or not, do they use the opening line or the closing line?

#3 - Are NFL 1st Quarter Spreads always -0.5 / +0.5?

Let me be clear about what I meant:

I want to have as much of my roll invested at any given time as possible. But I specialise in a sport that's very limited in +EV bets and I may not find enough bets to cover my roll (at any given time). Using full kellys will utilise my roll more than a 0.5 Kelly in cases where there simply aren't enough +EV bets out there. So basically my question is- is there a case to be made for using full Kelly if your specialising in a sport where few +EV bets are available?

Again, it is Kelly. Not Kellys. Or Kellies. Or Kelly's. Or whatever else you conjure up.

That idea is very dumb. Don't do that.

*Sigh* I guess ill follow 0.5 then... speaking as a lazy person, this comes as very bad news to me.
But thanks f/y advice.

And because this is meant to be a low content thread i shall finish with:

Shameone he he (a south park impersonation of Michael Jackson)

Inzaghi,

Question to ask (imo) would be: What would be the reason(s) for .5 compared to full?

If you understood that, then you would see why your previous question seemed so inane. Having no further curiosity about the underlying why doesn't really speak to you getting the concept either.

Thremp may call me dumb here, I'm aware of that likelihood.

Just bet whatever kelly multiple gets you to 100% of bankroll per day. .5, 1, 3x, 8x

Yeah your right I'm probably not curious enough. I did do a bit of groundwork on my own though. I was introduced to Kellie's *snigers* by the last chapter of the mathematics of poker (Chen).
I raised a question here about why a reputable sports bettor (propplayer) was recommending a boring 1-2% staking strategy rather than Kelly. And then somebody responded, to paraphrase, "well no one uses Kelly. Everyone uses 0.5 Kelly."

I wasn't sure what he meant by 0.5 but I googled it and soon discovered it's meaning:

Bet half of what Kelly criterion recommends

I encountered some articles that talked about the logic behind this in great detail and I didn't bother to read any of them. But i was drawn to a particular chart with an inverted curve that looked a bit like a mountain. On it was a line highlighting the position of a 0.5 Kelly strategy. The position appeared to maximise return per ounce of risk (or the Sharpe ratio as financiers would say). After studying it for a minute or two I was convinced (or as convinced as I was ever going to be without breaking too much of a sweat) that this was probably the best staking strategy. The correspondence I received here has somewhat confirmed that. God that was a boring story.

Anyway there was a practical point behind the question I posed- if you use full Kelly you don't have to do as much work because you don't have to find as many +EV bets as you would using 0.5 (all other factors constant). But I was told a few posts above that this is a stupid idea. So i accept that answer and shall use 0.5.

I thought that was all a waste of writing since he seemed incredibly lazy (dumb? Probably the same thing), but it took me far longer to figure out than it should have.

But ya, running a few dozen simulations is the easiest way to learn about this. Or even just discussing the results with other people? I've posted stuff in the past that I now know is totally wrong. Science or smth.

why can't it be Kelly's criterion? John Kelly created it so i feel like it's fair game to grant him the possessive form. unless he had multiple criterions, maybe that could get confusing. we could still call it Kelly's Kelly Criterion though.

i call it daliman's kelly criterion

shout out to 2007 2p2 wong teasers

yeah, but why can't it just be daliman criterion?

btw dailman sucks on twitter

didn't he end up being some sort of fezzik disciple or something?

Yea something like that. Hes ultra pompous.

he popped up on BJ21.com talking about how great he was or something and when people were like "yo who's the anonymous dude who thinks he's gods gift" he was like I'm daliman from poker and you guys need to quit saying fezzik isn't a winner because i've worked with him and it's great

How would I work out the combined odds of two dependent events in a football match?

For example:

Win + both teams to score is priced at 15/2
Win to nill is priced at 13/2

What are the combined odds? Obviously both events can't happen so it's effectively the same as betting on the team to win.

Thanks in advance if anyone can help.

thremp how much do you lol at the weekly NFL betting thread

Well, as you yourself suggest, combining the odds will simply reflect the odds of the team in question winning.

Prob (Team A wins) = Prob (Team A wins and Team B does not score) + Prob (Team A wins and Team B scores)

So, using the odds you provide:

Prob (Team A wins) = 2/17 + 2/15 *
= 64/255

*We can do this because the 2 events form a partition of "Team A wins".

Wait do I have this right.. RickySteve got butthurt over getting called out for being a supplicant beta cuck so he quit the forums a while back, and now came crawling back just to troll Thremp and post cat meme? Poor guy.