On the use of more-aggressive ramps to compare systems

[scroogemcduck]

Splitting this out of the CAC2 thread.

Well then, that's not a SCORE and not a meaningful comparison.

It’s the SCORE column straight out of CVCX. Just using a custom more-aggressive betting ramp to max out at 1.4% instead of 2.5% because table limits and heat are usually more constraining than bankroll in my experience. The custom betting ramps reduce the SCORE gap between REKO-F-6D and CAC2 from 13% to 8% on 5/6 S17 DAS LS RSA. This seems like a meaningful comparison if you’re often playing in places where your top bets are constrained by table limits and/or heat. Which is every counter eventually if he’s successful enough. I have a 2 million bankroll so there aren’t a lot of places I can go and actually use the “optimal” ramp. The best benchmark for comparing systems is CE/hour in the real world circumstances in which the system will be used by the individual player. Which is not perfectly correlated to SCORE, but also not similar enough between players to be a basis for public comparisons between systems. if you standardize bankroll=10k, unit size to set ror=13.5%, and RPH to 100, then CE/hour=SCORE/2

[BJGenius007]

Splitting this out of the CAC2 thread.



It’s the SCORE column straight out of CVCX. Just using a custom more-aggressive betting ramp to max out at 1.4% instead of 2.5% because table limits and heat are usually more constraining than bankroll in my experience. The custom betting ramps reduce the SCORE gap between REKO-F-6D and CAC2 from 13% to 8% on 5/6 S17 DAS LS RSA. This seems like a meaningful comparison if you’re often playing in places where your top bets are constrained by table limits and/or heat. Which is every counter eventually if he’s successful enough. I have a 2 million bankroll so there aren’t a lot of places I can go and actually use the “optimal” ramp. The best benchmark for comparing systems is CE/hour in the real world circumstances in which the system will be used by the individual player. Which is not perfectly correlated to SCORE, but also not similar enough between players to be a basis for public comparisons between systems. if you standardize bankroll=10k, unit size to set ror=13.5%, and RPH to 100, then CE/hour=SCORE/2

That is c-SCORE or custom-SCORE called for.

[Norm]

CVCX SCORE is only accurate if you don't alter the ramp. But, do what works for your circumstances.

[Cacarulo]

Splitting this out of the CAC2 thread.



It’s the SCORE column straight out of CVCX. Just using a custom more-aggressive betting ramp to max out at 1.4% instead of 2.5% because table limits and heat are usually more constraining than bankroll in my experience. The custom betting ramps reduce the SCORE gap between REKO-F-6D and CAC2 from 13% to 8% on 5/6 S17 DAS LS RSA. This seems like a meaningful comparison if you’re often playing in places where your top bets are constrained by table limits and/or heat. Which is every counter eventually if he’s successful enough. I have a 2 million bankroll so there aren’t a lot of places I can go and actually use the “optimal” ramp. The best benchmark for comparing systems is CE/hour in the real world circumstances in which the system will be used by the individual player. Which is not perfectly correlated to SCORE, but also not similar enough between players to be a basis for public comparisons between systems. if you standardize bankroll=10k, unit size to set ror=13.5%, and RPH to 100, then CE/hour=SCORE/2

I don't quite understand what you want to do, but a maximum limit could be solved by playing more hands. If we talk about SCORE, we have to discuss optimal bets with optimal betting ramps and a Risk of Ruin (ROR) of 13.53%.
Any change you make in the ramp will alter the ROR. If you bet aggressively (as you are suggesting), that ROR will be much higher. In the case of CAC2, your maximum bet only appears at a TC greater than or equal to +11,
and the advantage at that moment is about 4%.

Sincerely,
Cac

[scroogemcduck]

CVCX SCORE is only accurate if you don't alter the ramp. But, do what works for your circumstances.

By "not accurate" you mean it takes liberties with the definition of SCORE from Don's book. But custom-SCORE is still a useful metric for comparing one betting/wonging/playing strategy to another. I use it all the time. It still represents the EV/100 when betting full kelly on a 10k roll if the bets were scaled down proportionally.

In this toy example, customizing the betting ramp to be more aggressive is reducing my SCORE from 83.91 to 66.06, increasing my ROR from 0.0% to 0.0%, and increasing my hourly from 2037 to 3108.
Example.jpg

[Norm]

Yes, SCORE in its purist form is not used with an essentially infinite bankroll. A large team may not use SCORE in the pure form. But, as you say, C-SCORE is still useful.

[scroogemcduck]

I don't quite understand what you want to do, but a maximum limit could be solved by playing more hands. If we talk about SCORE, we have to discuss optimal bets with optimal betting ramps and a Risk of Ruin (ROR) of 13.53%.
Any change you make in the ramp will alter the ROR. If you bet aggressively (as you are suggesting), that ROR will be much higher. In the case of CAC2, your maximum bet only appears at a TC greater than or equal to +11,
and the advantage at that moment is about 4%.

Sincerely,
Cac

I use the formula 420/(playerhands+1) to approximate RPH.
After adjusting for the RPH effect, going from 2 hands of table max to 3 hands of table max only increases hourly by a factor of about (3/4)/(2/3) = 1.125.
Going from 3 hands of table max to 7 hands of table max only increases hourly by a factor of about (7/8)/(3/4) = 1.166

The hourly gain from a more aggressive ramp can be 50%+, which is much larger than the gain of going from 2 hands to 7.

[G Man]

I use the formula 420/(playerhands+1) to approximate RPH.

This is interesting, where did you get that formula ?