[JonWal]

Most people say don't risk more than 5% of total bankroll, but how good is this? Is there a calculator out there I can use to figure out the math behind it?

[Sacrelicious]

Windows comes with an excellent calculator application.

[JonWal]

Windows comes with an excellent calculator application.

Yea but is there a formula to figure out how effective risk is depending on accuracy?

[Sacrelicious]

Kelly criterion, crucial to any bettor.

[chopperocker]

its only suggested. if you love a play, go big.

[JonWal]

Kelly criterion, crucial to any bettor.

Thanks

its only suggested. if you love a play, go big.

Well even on a great play I like limits and understanding the best efficiency.
<br>
<br>

[Sacrelicious]

Yea but is there a formula to figure out how effective risk is depending on accuracy?

The biggest variable is determining accuracy, or rather, "edge", theres a great deal of information on this on the web.

Start a spreadsheet and track bets very closely, and depending on your estimated edge, use kelly criterion for determing bet sizes.

[Sacrelicious]

Also, begin with small bets in order to amass a sample size that exceeds standard deviation, its your best approach in order to determine your "edge" so that you can begin to properly kelly bet.

[JonWal]

The biggest variable is determining accuracy, or rather, "edge", theres a great deal of information on this on the web.

Start a spreadsheet and track bets very closely, and depending on your estimated edge, use kelly criterion for determing bet sizes.

Also, begin with small bets in order to amass a sample size that exceeds standard deviation, its your best approach in order to determine your "edge" so that you can begin to properly kelly bet.

<br>
<br>

Thanks this is exactly what I was looking for!

[JonWal]

The Kelly Criterion Calculation seems too risky. A simpler way is to make maximum risk the same % as your current edge. It's adjustable so you can optimize based on accuracy, more conservative than Kelly, and easier to calculate.

[Sacrelicious]

The Kelly Criterion Calculation seems too risky. A simpler way is to make maximum risk the same % as your current edge. It's adjustable so you can optimize based on accuracy, more conservative than Kelly, and easier to calculate.

Incorrect, kelly stating is always the ideal, however the variable in the Kelly equation is your perceived "edge", most people overestimate thier edge a great deal which is why they improperly size thier bets.

Kelly is always ideal, but in order to determine "edge" you need a massive sample size, which is why if you are just getting started betting it is better to flat bet, say, 1%, and track your bets meticulously. Its only through amassing a large sample size that you can begin to identify your edge, 1% is a safe amount to bet in order to amass a large enough sample size.

Make sense?

[JonWal]

Incorrect, kelly stating is always the ideal, however the variable in the Kelly equation is your perceived "edge", most people overestimate thier edge a great deal which is why they improperly size thier bets.

Kelly is always ideal, but in order to determine "edge" you need a massive sample size, which is why if you are just getting started betting it is better to flat bet, say, 1%, and track your bets meticulously. Its only through amassing a large sample size that you can begin to identify your edge, 1% is a safe amount to bet in order to amass a large enough sample size.

Make sense?

Yea that makes sense. So what would you say is a large enough sample size?

[HUY]

Yea that makes sense. So what would you say is a large enough sample size?

Until your results are statistically significant. You know what that means, right?

[statnerds]

kelly is flawed from the get cuz you will never be able to accurately gauge you edge.

also 5% will leave you broke in short order

Until you understand the following two items, which just might be the same principle stated differently, you are fukked.

Variance

Gamblers Ruin

[JonWal]

kelly is flawed from the get cuz you will never be able to accurately gauge you edge.

also 5% will leave you broke in short order

Until you understand the following two items, which just might be the same principle stated differently, you are fukked.

Variance

Gamblers Ruin

Yea I understand variance. Basically if you flip a coin 100 times you might have 65% heads and 45% tails and the more flips you do will get closer to 50/50. So betting low is protects you from that variance by giving you more chances long term.

[JonWal]

Yea I understand variance. Basically if you flip a coin 100 times you might have 65% heads and 45% tails and the more flips you do will get closer to 50/50. So betting low is protects you from that variance by giving you more chances long term.

Lol *65% heads and 35% tails...

[Sacrelicious]

Yea that makes sense. So what would you say is a large enough sample size?

The more the better. I'm no professional, I'm actually quite a novice, I'm slowly in the process of going from flat betting to kelly betting, also depends on the sport.

I only bet MMA and gimmicky prop bets, they have the most value, for traditional stick and ball sports I'd say hundreds of samples.

Start flat betting 1%, as soon as you notice trends and begin to isolate an edge, start making the transition to kelly, or even half kelly.

[stevenash]

Bet coin flip situations where you can get + 125 or better on your wager.

Without blowing my own horn, I demonstrated this many times over last baseball season.

[shackfu99]

Yea that makes sense. So what would you say is a large enough sample size?

minimum of 300 wagers per subset.

[Sawyer]

Depends on your bankroll. %5 can be used only for small bankrolls such as 1000-2000$. For larger bankrolls, you need a more conservative %.

Some other factors are also in play though such as number of plays etc. How many games you bet/month? You're a SSSS? (Selective Sniper Sharp Shooter) Or you're a Volume Player (or tailing other cappers) who bets 20-30 games in a day. You decide.

My suggestion is,

Below 2,000 %5 is okay.
2,000-5,000 %4
5,000-10,000 %3
10,000-20,000 %2.5 or %3
20,000-50,000 %2 or %2.5
50,000-100,000 %1 or %2
100,000+ %1

[HUY]

Bet coin flip situations where you can get + 125 or better on your wager.

Without blowing my own horn, I demonstrated this many times over last baseball season.

How many times are those "many" times?

[stevenash]

^
Check my posting history

[sweep]

^
Check my posting history

Hey Nasher how but some freakkin' clif notes buddy
Dont feel like searching back to last bb season. PM if u want