Kelly value betting question

**danshan11** · 08-30-18, 11:41 PM

read this hope it helps!

https://www.pinnacle.com/en/betting-articles/Betting-Strategy/fractional-kelly-criterion/GBD27Z9NLJVGFLGG

**hcetigol** · 08-31-18, 01:01 AM

Thanks danshan11, but not really. All I want to know is; if the bettor already knows what they will average in terms of closing edge in the long run over thousands of bets, then wouldn't it be better to use that figure verses the edge figure at the time of placing, which can obviously increase or decrease at close (ultimately though will average out to 5 % as per the example given). Do you guys think this could be more profitable than using the edge figure at time of placing to calculate the stake? Cheers

**tsty** · 08-31-18, 02:51 AM

Originally posted by hcetigol

Thanks danshan11, but not really. All I want to know is; if the bettor already knows what they will average in terms of closing edge in the long run over thousands of bets, then wouldn't it be better to use that figure verses the edge figure at the time of placing, which can obviously increase or decrease at close (ultimately though will average out to 5 % as per the example given). Do you guys think this could be more profitable than using the edge figure at time of placing to calculate the stake? Cheers

no because every game is different

betting on average edge is stupid because some games you will have 3% and others 8%

bet depending on your model

that's it

**Alfa1234** · 08-31-18, 04:47 AM

Originally posted by tsty

no because every game is different

betting on average edge is stupid because some games you will have 3% and others 8%

bet depending on your model

that's it

This.

If your average edge is 5% you still have games in there with an edge of 15% and others with an edge of 2%. Use the sharp line as a base to calculate your bet %.

**danshan11** · 08-31-18, 08:18 AM

I would NEVER use kelly as it is I use a 3rd cousin, I call it fixed profit betting which is for sure a kelly cousin but not kelly as it is
if I did use Kelly I would base my edge on the actual line at pinnacle NOT my model or my guess or any long term average. The most accurate measure we have of a game is the current line and it only gets sharper at closing, Hope that helps
and if you care my fixed profit is bet whatever is necessary to win 1% of your bankroll on every wager
-200 bet 200 to win 100
100 bet 100 to win 100
200 bet 50 to win 100
this gives you more money on things more likely to happen and less money on things less likely to happen but honestly I backtested this on compared to flat betting and on about 500 bets I made like 2 or 3 more units with this method BUT the big advantage was it did not draw down as much in the rough times and that can be valuable mentally and on your bankroll

**tsty** · 08-31-18, 10:15 AM

Why do you think sharps move the line the way they do?

**danshan11** · 08-31-18, 10:33 AM

if you are asking me, I think they are trying to gain any gap in the current line to what they believe the line is suppose to be or will close at

**tsty** · 08-31-18, 10:41 AM

how do they accomplish this?

**danshan11** · 08-31-18, 11:08 AM

Originally posted by tsty

how do they accomplish this?

two main ways but I am sure there are more
1 they soften the line and then pound it
2 they bet it and rebet it and rebet it until it is what they think is fair

**Alfa1234** · 08-31-18, 06:17 PM

Originally posted by danshan11

two main ways but I am sure there are more
1 they soften the line and then pound it
2 they bet it and rebet it and rebet it until it is what they think is fair

What do you mean by "soften the line"?

**danshan11** · 08-31-18, 06:23 PM

they go a little dough on the other side move it a pinch and then come in and pound hard the number they like

**danshan11** · 08-31-18, 06:24 PM

most of the big guys give the wrong pick to their million subscribers wait for the action on it and bet the other side. They dont mess up their success overall but they dont mind squeezing their users for a game or so a week!

**tsty** · 08-31-18, 10:53 PM

But why do they choose the side they do?

There is obviously a reason right? They don't just bet in the hopes that their decision is correct right?

**Alfa1234** · 09-01-18, 07:17 AM

Originally posted by danshan11

they go a little dough on the other side move it a pinch and then come in and pound hard the number they like

This does not work as the line moves proportionally to the amount of money put on it. The only way it can work is by pounding 1 side when the line opens and bet higher limits once the limits go up. This is extremely risky though as anyone that can identify the real value will bet that line early as well and simply get more value.

Waiting for limits to go up has the same risk, someone who has identified the value will take it early. Low limits or not.

**danshan11** · 09-01-18, 08:19 AM

Originally posted by tsty

But why do they choose the side they do?

There is obviously a reason right? They don't just bet in the hopes that their decision is correct right?

that side has vallue
they model yanks -150 assume nothing changes and line is -120, they will bet the Yanks, they dont choose the side, the side chooses them based on the number. The reality is their bankroll and staking method determines how much they will adjust the line.

the real cause of yanks at -120 and not -150 to begin with is dumb people like me who think the opponent is on a streak, or starter is coming off a good start or bad start, things that dont really matter and push it the wrong way and the sharps come in and pounce on my bets. I learned that quickly on the exchanges, you put up a bet on the value side of a game its gone in an instant, you put up a bet on the non value side it sits for eternity.
what is the moral of these questions? if it is to test my betting IQ, there is no need I can tell you now, I am a nerf that has zero clue how to do this. I know nothing about math and less about bookmakers and why they do things. I have a real simple formula, bet a line that will move enough to cover the margin more than 50% of the time, that is all I try and do. Now this practice on paper is great but in reality with variance and my lack of ability to model over max I would say 100 games across a few leagues a day and only can find value in say 10% of those games, this gives me a tiny sample size and leaves me wide open to huge variance swings and the potential to never see my true edge because of the lack of total number of bets.

**jbayko** · 09-01-18, 11:12 AM

Originally posted by Alfa1234

This does not work as the line moves proportionally to the amount of money put on it. The only way it can work is by pounding 1 side when the line opens and bet higher limits once the limits go up. This is extremely risky though as anyone that can identify the real value will bet that line early as well and simply get more value.

Waiting for limits to go up has the same risk, someone who has identified the value will take it early. Low limits or not.

Right. You can't get everyone to cooperate with you, and in order for this to work, you'd need everyone to cooperate. I could see it maybe working in smaller niche markets, but that's why books add lots of extra protection in those markets, including higher juice lower limits, and fast limiting of players.

**danshan11** · 09-01-18, 12:50 PM

Originally posted by Alfa1234

This does not work as the line moves proportionally to the amount of money put on it. The only way it can work is by pounding 1 side when the line opens and bet higher limits once the limits go up. This is extremely risky though as anyone that can identify the real value will bet that line early as well and simply get more value.

Waiting for limits to go up has the same risk, someone who has identified the value will take it early. Low limits or not.

i think you are right but at the same time, I think syndicates give out bad info on one out of so many and pound the opposite after they release the setup pick. Look at places like RAS say they wanted to really bet the ranch on Texas -7 they would give out Ohio +7 and it would move to -5 on that play and they could come back and pound Texas at -5 and enjoy 2 points that wont be there once the market adjusts back. I am not saying RAS does that, I am saying it is very easy and possible. but your point is valid and I dont even care about this conversation, I am not sure why tsty even asked me these questions honestly.

**danshan11** · 09-01-18, 12:51 PM

Originally posted by tsty

But why do they choose the side they do?

There is obviously a reason right? They don't just bet in the hopes that their decision is correct right?

why are you asking me this?

**HeeeHAWWWW** · 09-01-18, 03:59 PM

Originally posted by danshan11

i think you are right but at the same time, I think syndicates give out bad info on one out of so many and pound the opposite after they release the setup pick.

The likes of RAS, nah. It certainly happens with fraud tipsters on social media though. They get affiliate profit share for the losses they generate, then make double money by betting the other side.

**danshan11** · 09-01-18, 05:18 PM

yes agree, I am not saying RAS i am just saying they have that capability

**tsty** · 09-01-18, 10:49 PM

ras are garbage lol

**danshan11** · 09-02-18, 09:39 AM

I am not a big RAS fan, If your interested let me know and I will give you my full opinion on them.

**tsty** · 09-02-18, 11:27 AM

ras hasnt had a winning month in the last 2 years

just another scum bag tout

**danshan11** · 09-05-18, 10:08 PM

do they lie or how are you saying that? they claim to win every year on their site???