Quote:WizardQuote:ssho88Based on your defined game rules and average multiplier values, as well as conventional basic strategy, my 1 billion trials simulation results show that the house advantage is 18.09%. A little bit different from yours, I don't know where I went wrong.

I'm actually happy we are that close. This is a difficult game to simulate as there are lots of different game states. It is possible the error is on my end too.Quote:The current hand strategy largely depends on (current hand EV1 + next hand EV2), multiplier has a great influence on the basic strategy, and the correct basic strategy requires more accurate multiplier values.

I agree.Quote:Another question : Suppose you split, lose one hand and win the other. Is your multiplier still active in the next round ?

link to original post

I interpret the answer to that question as yes. You need just one win to earn a multiplier. Somebody correct me if I'm wrong.

link to original post

I found that in your Expected Return Table(https://wizardofodds.com/games/lightning-blackjack/), the probability of DOUBLE WIN and DOUBLE LOSS is slightly higher than my figures. I guess we are using different conventional basic strategy, should HIT for 11 vs T instead of DOUBLE ?

They don't peek on Dealer ten. Though Wizard says it makes no difference because one only loses original wager, I reckon you should NOT double on 11 vs Dealer 10Quote:ssho88

I found that in your Expected Return Table(https://wizardofodds.com/games/lightning-blackjack/), the probability of DOUBLE WIN and DOUBLE LOSS is slightly higher than my figures. I guess we are using different conventional basic strategy, should HIT for 11 vs T instead of DOUBLE ?

link to original post

https://wizardofvegas.com/forum/questions-and-answers/advice/36630-evolution-lightning-blackjack/5/#post830953

What a brilliantly sneaky game this is. Looks like Blackjack, Plays like blackjack. 18% house edge if you play Blackjack strategy. Who would expect that?

Optimum strategy must be something like 'Maximise probability of a win to lock in the bonus', NOT 'maximise the EV on the base game'

That might even mean you should sacrifice classic +EV plays in favour of a higher probability, smaller win.

Whatever strategy changes can make so much difference?

Roll on with best strategy, someone?

Quote:OnceDearThey don't peek on Dealer ten. Though Wizard says it makes no difference because one only loses original wager, I reckon you should NOT double on 11 vs Dealer 10Quote:ssho88

I found that in your Expected Return Table(https://wizardofodds.com/games/lightning-blackjack/), the probability of DOUBLE WIN and DOUBLE LOSS is slightly higher than my figures. I guess we are using different conventional basic strategy, should HIT for 11 vs T instead of DOUBLE ?

link to original post

https://wizardofvegas.com/forum/questions-and-answers/advice/36630-evolution-lightning-blackjack/5/#post830953

link to original post

Correct, I just HIT 11 vs T. May be that is the reason why my probability of DOUBLE WIN/LOSS is lower than WIZARD's figures.

Quote:OnceDearWouldn't it make a strategy difference? I.e. Peek on ten, you double an ace against 10, knowing he doesn't have blackjack. No peek, you would hit rather than double an ace against ten?Quote:Mission146RE: Peek Policy

Do you happen to know, based on the dealer not peeking when a ten is up, whether the player loses the full amount bet or just the original bet to a dealer natural? As I understand it, if the player would lose the original bet only, then it doesn't mathematically matter whether or not the dealer peeks.

link to original post

link to original post

That's why it matters whether or not you only lose the original bet. If it's no peek, but you double and lose, but it's also only lose the original bet...then the amount that you doubled for gets pushed back to you.

Quote:ssho88I found that in your Expected Return Table(https://wizardofodds.com/games/lightning-blackjack/), the probability of DOUBLE WIN and DOUBLE LOSS is slightly higher than my figures. I guess we are using different conventional basic strategy, should HIT for 11 vs T instead of DOUBLE ?

link to original post

I should have had the player hit 11 vs. 10, but I now see that I forgot to make that change. My bad. I don't think I will rerun the simulation over just this. I don't want to invest too much time into this if I don't have exact numbers for the average multipliers.

Quote:WizardI should have had the player hit 11 vs. 10, but I now see that I forgot to make that change. My bad. I don't think I will rerun the simulation over just this. I don't want to invest too much time into this if I don't have exact numbers for the average multipliers.

link to original post

After changing the strategy of 11 vs T to DOUBLE, my simulation results match all the numbers in your Expected Multiplier Table and Expected Return Table.

Even if you know the exact multiplier values, it is difficult to find correct basic strategy. When you change the strategy, the TOTAL EXPECTED MULTIPLIER and PROBABILITY OF HAND OUTCOME will change, this will affect your EXPECTED RETURN, then it will affect your stretegy.....I guess must use trial and error method to find correct basic strategy, what do you think ?

Quote:ssho88After changing the strategy of 11 vs T to DOUBLE, my simulation results match all the numbers in your Expected Multiplier Table and Expected Return Table.

link to original post

Thank you! I think I will rerun it again, out of respect for you.

If the multipliers are randomised at each round, is it true to say that the optimal strategy will change for every hand. That would make it impossible for a human to play optimal strategy. The 99.56% would be a pipe dream.Quote:ssho88Quote:WizardI should have had the player hit 11 vs. 10, but I now see that I forgot to make that change. My bad. I don't think I will rerun the simulation over just this. I don't want to invest too much time into this if I don't have exact numbers for the average multipliers.

link to original post

After changing the strategy of 11 vs T to DOUBLE, my simulation results match all the numbers in your Expected Multiplier Table and Expected Return Table.

Even if you know the exact multiplier values, it is difficult to find correct basic strategy. When you change the strategy, the TOTAL EXPECTED MULTIPLIER and PROBABILITY OF HAND OUTCOME will change, this will affect your EXPECTED RETURN, then it will affect your stretegy.....I guess must use trial and error method to find correct basic strategy, what do you think ?

link to original post

Quote:OnceDearIf the multipliers are randomised at each round, is it true to say that the optimal strategy will change for every hand. That would make it impossible for a human to play optimal strategy. The 99.56% would be a pipe dream.

link to original post

We should record the multipliers of hundreds/thousands of hands, and then only apply a fixed set of average multipliers.