How does bet spread affect win rate

[qzj]

I am wondering how a 1-10 bet spread differ from 1-20 bet spread in terms of win rate per hour. PBJ suggested a benchmark of 1-10 bet spread where at 0 you bet 10, $25 at 1, $50 at 2, $75 at 3 and $100 at 4+, whereas when I read Kevin Blackwood's book, he suggested a 1-20 bet spread where at 0 you bet $5, $20 at 1, $40 at 2, $60 at 3, $80 at 4 and $100 at 5+.
I could see that Wong's benchmark would reach your maximum bet faster and relatively more frequently than Kevin's approch. But how will this difference be reflected on the win rate and many other factors? Many thanks.

[rollem411]

Simply put, you are betting 10$ at a disadvantage opposed to 5$ which would lower your winrate.

[Kasi]

Originally Posted by qzj

I am wondering how a 1-10 bet spread differ from 1-20 bet spread in terms of win rate per hour. PBJ suggested a benchmark of 1-10 bet spread where at 0 you bet 10, $25 at 1, $50 at 2, $75 at 3 and $100 at 4+, whereas when I read Kevin Blackwood's book, he suggested a 1-20 bet spread where at 0 you bet $5, $20 at 1, $40 at 2, $60 at 3, $80 at 4 and $100 at 5+.
I could see that Wong's benchmark would reach your maximum bet faster and relatively more frequently than Kevin's approch. But how will this difference be reflected on the win rate and many other factors? Many thanks.

Each different spread would give rise to a different avg bet size and a different overall advantage percentage. Even assuming the exact same game and same everything so that the advantage and frequency associated with each count is the same.

If you want to maximize earnings per unit of time, hands played per hour enters into it. I'm assuming that's what you're talking about when you say win rate.

So, by itself the overall advantage, avg bet size and hands played per hour don't really mean that much until you multiply them together and see which one makes you more money in an hour.

And if you also care about stan dev, rather than choosing which game to play based soley on maximizing win rate, that could change too.

[qzj]

Kasi, I appreciate your reply very much. Sorry if the term wasn't clear, by win rate, I do mean the earnings per hour. I am assuming that holding all the other factors constant(rules, number of hands per hour), how would a bigger bet spread change your win rate? I am thinking that Wong's 1-10 bet spread reaches $100 at 4+, whereas Kevin's 1-20 bet spread reaches $100 at 5+, so natually, 1-10 bet spread has a higher chance of winning more per hour since you are betting bigger more frequently than a 1-20 bet spread, but your risk of losing more dollars per hour has increased as well, and hence std dev is increased since your wins and losses would spread further apart from the mean. If a larger spread would increase your standard deviation, why do pros suggest a larger bet spread for shoe games, and a smaller bet spread for single or double deck games? Is there some kind of reasoning behind this?

[Kasi]

Originally Posted by qzj

how would a bigger bet spread change your win rate? I am thinking that Wong's 1-10 bet spread reaches $100 at 4+, whereas Kevin's 1-20 bet spread reaches $100 at 5+, so natually, 1-10 bet spread has a higher chance of winning more per hour since you are betting bigger more frequently than a 1-20 bet spread,

It's possible the bigger spread could have a lower avg bet but still have a higher EV because the overall win %age is that much greater than the 1-10 with a higher avg bet but a lower overall advantage.

Probably anything is possible lol - it just depends on how much you bet at each TC, the frequency of that TC and the advantage that each TC gives rise to. And I guess variance could be whatever also depending on how many units you bet and when.

Like in your example the $5 bet might represent a 2000 unit roll but the $10 bet only a 1000 unit roll.

I guess that's why the easy answer is use a sim because that's the only way you can get the frequencies of the TC's and the advantage at each TC. The variance too at each TC but that doesn't change a whole lot. But it does change. You know nothing until you know those 3 things. If you do know them, the keys to the kingdom are yours lol.

Like in your example, since you're assuming TC's frequencies and advantages and variance at each TC is identical, if you don't know what they actually are, make some reasonable guesses and see what changes for the fun of it when you multiply the freq and advantages and variances (1.32 is a reasonable number maybe). You should be able to get avg bet, overall adv, SD, and a bunch of other stuff.

Hope this helps a little.

[sagefr0g]

Originally Posted by qzj

Kasi, I appreciate your reply very much. Sorry if the term wasn't clear, by win rate, I do mean the earnings per hour. I am assuming that holding all the other factors constant(rules, number of hands per hour), how would a bigger bet spread change your win rate? I am thinking that Wong's 1-10 bet spread reaches $100 at 4+, whereas Kevin's 1-20 bet spread reaches $100 at 5+, so natually, 1-10 bet spread has a higher chance of winning more per hour since you are betting bigger more frequently than a 1-20 bet spread, but your risk of losing more dollars per hour has increased as well, and hence std dev is increased since your wins and losses would spread further apart from the mean. If a larger spread would increase your standard deviation, why do pros suggest a larger bet spread for shoe games, and a smaller bet spread for single or double deck games? Is there some kind of reasoning behind this?

back to the 'good ole book' again might shed some light on this matter:
http://www.blackjackincolor.com/penetration5.htm

[qzj]

Originally Posted by Kasi

It's possible the bigger spread could have a lower avg bet but still have a higher EV because the overall win %age is that much greater than the 1-10 with a higher avg bet but a lower overall advantage.

Make sense! Thanks for clearifying this for me.


To safefrog: great link, thanks man.