Where can I find a double down for less strategy?
- Thread starter SilentBob420BMFJ
- Start date
SilentBob420BMFJ
Banned
i dont understand why tho, because think about it, if you have lets say a 49% chance of winning a hand, why would you put more money out there? im pretty sure the other dude was right in that you have a greater than 50% chance of winning all double downs, thats why you double.. in fact, how is it possible that a double down has a higher chance of winning the hand than hitting? correct me if im wrong, but a double down can never have a higher chance of winning compared to hitting, but if its like 63% vs 64%, its best to double even tho you have a lower chance of winning the hand, because 63% of $20 is better than 64% of $10Kasi said:
ya, this was from a scavenger topic, but it has nothing to do with not having the money, its a complex article but it made sense, because when you double another persons hand, you dont own the original bet, thus you are doubling down for free basically.. pretend the dealer dealt cards to an empty spot, and then after the cards were dealt, you could bet on that spot, thats essentially what scavanger doubling is, and i forgot all the details but it worked for splitting too, and there was a splitting/doubling fractions chart.. oh wait i still have it, here it is.. http://www.blackjackforumonline.com/content/scavengerbjfo.pdf scavengerbjfo.pdf (application/pdf Object)GeorgeD said:
Sonny
Well-Known Member
You’re absolutely right. Since you only get one card on a double down the probability of winning is usually smaller than hitting. The only exception is a situation where you would only take one card anyway, like 11 vs. 3-6. Even if you hit you would never take more than one card so the probability of winning is the same for hitting and doubling.SilentBob420BMFJ said:
Because you are getting paid more money. See below.SilentBob420BMFJ said:
You’ve just answered your own question. Think about it. If hitting will win 50% of the time but doubling only wins 30% of the time (but pays double), the EVs are:SilentBob420BMFJ said:
.5 * $10 = $5
.3 * $20 = $6
You win more money by doubling even though the probability of winning is much lower.
No, the probability of winning can be far less than 50% and you can still have a +EV situation because you are getting 2:1 odds.SilentBob420BMFJ said:
-Sonny-
I don't think this is correct. You need a greater than 50% chance of winning your double to have a positive EV. Say you know you have an exactly 50% chance of winning a double. Your doubling EV (=0) can't be more than your hitting EV (>=0) since hitting will have at least a 50% chance of winning, maybe more if hit correctly. As long as your double succeeds more than 50% of the time, the potential for increased EV by doubling instead of hitting exists. The general rule is that if hitting wins (50 + x) percent of the time, doubling must win more than (50 + x/2) percent of the time to make doubling EV greater than hitting EV, assuming you exactly double your wager.Sonny said:
Now take a case where doubling yields a winning percent of less than 50%, for example 45%. A 45% winning rate on a doubled bet is a -20% EV. If instead you hit this hand, you would need to win 40% of the time to have a -20% EV. However, you would never have an winning percent of less than 45% on the hit hand unless the hand is misplayed, so the worst case hitting EV would be -10%. All you are doing by doubling a hand with a winning percent of less than 50% is getting a more negative EV.
I don't know if I explained it very well.
There is a case where you may win less than 50% of the time. This is when the assumption is player loses to dealer's blackjack. For example, 5-6 versus ace, single deck, full peek: hit EV=-16.47, double EV=-21.01. However, player does win more than 50% if dealer's blackjacks are excluded: hit EV=+17.29, double EV=+24.03. Using either set of figures, the correct strategy is apparent. What I'm trying to show is that even though player wins less than 50% overall, he still wins more than 50% once it is known dealer doesn't have blackjack, so it still agrees with what I said above.
k_c
Sonny
Well-Known Member
It all depends on the probability of a tie as well. You’re right that my example above isn’t exactly applicable to BJ because it does not take into account the losing side of the EV, but a BJ bet can still be +EV even with less than a 50% probability of winning. For example, imagine a bet with only a 47% chance of winning, but also an 8% chance of a tie. The EV looks like this:k_c said:
=.47*20 + .08*0 + .45*-20
=9.4 + 0 - 9
=0.4
We have a +EV with a low probability of winning. But, that being said, your understanding of the hit/double decision is completely correct. The only difference is that you should look at the EV, not just the win probability.
-Sonny-
You can resolve any discrepancy by counting pushes as 50% wins and 50% losses.Sonny said:
.51*20 + .49*(-20)=0.4
(Edited to be consistent with previous example)
Another way given win rate counting ties as 50% W, 50% L to compute a double:
(2*win rate - 1)*2(since bet is doubled)*(initial bet)
=(2*.51-1)*2*10=0.4
It's all good.
k_c
Last edited:
Maybe a picture is worth a 1000 words lol.
Anyway, I always felt better with specific examples rather than thinking purely theoretically. I could always understand addition and subtraction alot easier than general formulae lol.
So this is my reconciliation of the EV of doubling a 5,4 vs 3 in a 6D,S17 game.
If one doubles, one will win 49.9%, push 6.868%, and lose 43.233% of the time.
Therefore one's EV via doubling, assuming 1000 hands at $1, is $2000*.499=998. Less $2000*.43233=864.66.
998-864.66=133.34.
Compare to the Wiz's EV of doubling a 5,4 vs 3 of 0.133343 in a 6D,S17 game.
Likewise for 7,3 vs 9 with W 49.3%, P 8.689% L 42.011% doubling percentages, EV=1457.8 compared to Wiz EV of 0.145776.
So I hope this shows that the max EV is realized even though one will win fewer than 50% of the total hands dealt when doubling.
I'll let anyone else reconcile in a similar way the EV's that the Wiz gets for hitting the same 2 hands.
Remember that, in backgammon, no ties allowed lol, always accept the doubling cube even if you will lose 74% of the time lol.
Sonny - I agree with what you say, that the EV of doubling would be exactly twice the EV of hitting 11 vs 4-6 but maybe not for the 11 vs 3. No big deal -we all know what u meant.
Also, in case it matters to anyone, I always think of EV as a % of average original wager per dealer upcard, not average total wager per dealer upcard.
But, really, I was much more interested in how BS would change if one were allowed to double for one more card after splitting Aces. Not that I'm likely to get to Northern Michigan lol.
Anyway, I always felt better with specific examples rather than thinking purely theoretically. I could always understand addition and subtraction alot easier than general formulae lol.
So this is my reconciliation of the EV of doubling a 5,4 vs 3 in a 6D,S17 game.
If one doubles, one will win 49.9%, push 6.868%, and lose 43.233% of the time.
Therefore one's EV via doubling, assuming 1000 hands at $1, is $2000*.499=998. Less $2000*.43233=864.66.
998-864.66=133.34.
Compare to the Wiz's EV of doubling a 5,4 vs 3 of 0.133343 in a 6D,S17 game.
Likewise for 7,3 vs 9 with W 49.3%, P 8.689% L 42.011% doubling percentages, EV=1457.8 compared to Wiz EV of 0.145776.
So I hope this shows that the max EV is realized even though one will win fewer than 50% of the total hands dealt when doubling.
I'll let anyone else reconcile in a similar way the EV's that the Wiz gets for hitting the same 2 hands.
Remember that, in backgammon, no ties allowed lol, always accept the doubling cube even if you will lose 74% of the time lol.
Sonny - I agree with what you say, that the EV of doubling would be exactly twice the EV of hitting 11 vs 4-6 but maybe not for the 11 vs 3. No big deal -we all know what u meant.
Also, in case it matters to anyone, I always think of EV as a % of average original wager per dealer upcard, not average total wager per dealer upcard.
But, really, I was much more interested in how BS would change if one were allowed to double for one more card after splitting Aces. Not that I'm likely to get to Northern Michigan lol.
6D, S17, DA2, SPL3, DAS, HSA, RSA to 4 hands, double after split acesKasi said:
Overall EV = -.1615%
My program assumes that in order to double after split aces, both DAS and HSA (hit split aces) must be allowed, so it may not be exactly what you're looking for, though.
k_c
Thanks k_c I appreciate it.k_c said:
And I'm not sure if it helps me or not lol.
I was just wondering if I split AA, and get, say, a 2 on one ace and a 5 on the other ace, and my only choice is to either stand every time (normal rules) or double my bet and get one more card, I'm thinking I would be better off doubling vs every dealer upcard. Even vs 10's and aces.
Not sure I see the point of doubling for less, even if so.
Just trying to determine the correct BS for anyone who may actually play this game so they realize the max increase in HA due to this rule.
I guess I'd guess most would only double according to normal BS.
And, of course, I could also be nuts in suggesting this lol.
But, if I'm right, I may just travel to Northern Michigan to really piss people off when I double my A,5 vs A after splitting Aces lol.
One card draw to split ace hand
A-5 versus ace:
Stand: -66.54%
One card draw, doubled bet: -62.93%
One card draw, undoubled bet: -31.46%
I can get this type of data, just not an overall EV. The overall EV can't be better than -.1615% for typical 6D, s17, split to 4 hands. Your best strategy on this hand would be to double for less for as little as possible (zero, if possible.)
k_c
Off the top of a 6D shoe:Kasi said:
A-5 versus ace:
Stand: -66.54%
One card draw, doubled bet: -62.93%
One card draw, undoubled bet: -31.46%
I can get this type of data, just not an overall EV. The overall EV can't be better than -.1615% for typical 6D, s17, split to 4 hands. Your best strategy on this hand would be to double for less for as little as possible (zero, if possible.)
k_c
Cool!k_c said:
Is the "one card draw doubled bet: EV of -62.93%" expressed as a percent of original bet or total bet?
In any case, am I reading it right that one loses less by doubling than standing? Since there will be no choice of hitting the hand without doubling after they are split in this game?
Also, I suppose, at least one more ace would always be gone since they are split. Not sure what difference that makes.
In any case I think A,5 vs A might be the most marginal hand.
SilentBob420BMFJ
Banned
yes, thats right, but as you can see, doubling is a horrible play, its just better than standing, which shows why HSA is much better than RSAKasi said:
most marginal compared to what?
anyways, its quite obvious that doubling is better than standing in any situation where hitting is much much better than standing
It's relative to initial bet. The loss would be 31.46% if one unit is bet, but since an additional unit is added when doubling for the full amount, the loss EV is doubled.Kasi said:
If the only choices are standing or doubling for the full amount, doubling is better.Kasi said:
I just gave the figures for a full shoe as a quick answer. You are right in saying that at least one more ace would be removed. If the A-5 versus A occurred on the first split hand, at that point 3 aces and one five would be removed:Kasi said:
stand: -66.47%
full double: -62.88%
one card undoubled: -31.44%
Splits can be complicated to compute, but that should show the general picture.
k_c
The point here was to establish basic strategy in a game where one is allowed to double or not one's 2-card hand after splitting aces. Not whether HSA is better than RSA. In an HSA game one is not limited to hitting only once.SilentBob420BMFJ said:
Since one loses less by doubling an A,5 vs Ace than always standing on it, the basic strategy play would be double that hand, as k_c established.
Generally speaking, I don't usually call the best play available, the basic strategy play, a "horrible" play.
That it's the hand that yields the least benefit to the BS player compared to standing. The BS here being to double any A2 thru A5 vs dealer anything in a game where one is allowed to double a hand and receive one card after splitting Aces.SilentBob420BMFJ said:
I'm not totally sure it's true but that what's I meant. I guess doubling A,2 vs 2might be at the other extreme.
Well, when we get around to the basic strategy of what to do with an A,6 vs any dealer upcard in this game, it sounds like, since it's so obvious to you, that you would double an A,6 vs a dealer 10 or Ace, just as you would an A,5, since hitting an A,6 vs a 10 or Ace is better than standing on it.SilentBob420BMFJ said:
Is that what you would do in a 6D S17 game where you are allowed to double a 2-card hand and receive one card after splitting Aces?
Or is that a situation where hitting is not "much much better than standing"?
Maybe it's only a one "much" better situation?
Mr. Obvious, (he is is a funny radio character here so no offense intended), exactly what would be your BS in this game when you receive an A,6 after splitting Aces vs each dealer upcard?
Hint, I said 6 days ago what I thought it was. You agree or disagree?
On another subject, are you at least cool with the fact, at least I think it's a fact, since you never jumped in with what you thought after I said what I thought, that sometimes, even though one will win less than 50% of doubled hands, it may still be the right play? I think you also thought then that it was obvious such a thing would make no sense and not be possible.
After waking up with a clear mind, (trust me, if it's after midnight, there's a greater chance that 6-5 is better than 3-2 than I'm thinking straight), I completely understood.k_c said:
Thanks for confirming, even at a full double, it still would be the BS play.
The most interesting thing to me, and very much thanks to you and miplet for getting me thinking in different ways than I am used to, I think we, well, OK, you and miplet, have established that the traditional wisdom of "if it's correct to double, it's never correct to double for less", is simply not correct under all rules.
At least in a 6D S17 game where one is allowed to double after splitting Aces.
And probably also a 4-8D S17 game.
Since we are dealing with negative EV's in the first place.
Quite an epiphany. For me anyway.
Just another example, and by no means the first lol, of what I once thought was "absolutely true", just ain't so.
You know it may just well be worth the trip to Northern Michigan, just to double for 25 cents on a $10 bet with an A,5 vs A and incur the wrath of those whose "card I stole" while being such an obvious idiot
I think I have a pretty good idea of the hands it's better to "double for less" on, even though it still would be OK (BS) to double for the full amount on, but, if you, or miplet, or anyone, feel like going above and beyond the call of duty, the more so since if I said what I thought nobody would believe it anyway, including me lol, and confirm the specific hands that it would be correct to "double for less" on, or, indeed, confirm BS for all hands even if one doubled for the full amount, including A,6, for any who actually live in Northern Michigan and play this weird game I never heard of, well, then, God bless you.
And I thought William Faulkner had dense sentences lol.
6 Decks s17 Where you may get a third card after Splitting Aces by Doubling down. Hitting not an option.
F= Always double for the full amount
S= Always stand
L= Double for less if aloud, otherwise double the full amount
#=Stand unless you can double for less the # shown based off a $100 bet.
Hope I didn't make any errors. Use at your own risk.
Code:
A 2 3 4 5 6 7 8 9 10
A,A L L F F F F L L L L
2,A L L L F F F L L L L
3,A L L L F F F L L L L
4,A L L L F F F L L L L
5,A L L L F F F L L L L
6,A 78.673 L F F F F L L L 84.59
7,A S S F F F F S S 28.247 4.778
8,A S S S S S S S S S S
9,A S S S S S S S S S S
10,A S S S S S S S S S SS= Always stand
L= Double for less if aloud, otherwise double the full amount
#=Stand unless you can double for less the # shown based off a $100 bet.
Hope I didn't make any errors. Use at your own risk.
Last edited:
Thanks miplet. Great work.miplet said:
If I ever play this game and lose money, I'm coming back at you :grin:
I never would have gotten the double for less amounts on some of those hands.
And probably not some of those hands like A,2 vs 4 where u say to double for full amount (once I figured out that doubling for less is often the best play lol).
Doubling A,7 vs 9 and 10 for less?!. Beautiful. Everyone will hate me for sure. This is a game I'd enjoy playing lol.
Where is this game again? I get to southern Michigan off and on.
Any chance, and I hate to even ask and put you to more work, but do you have any idea of the overall benefit to a player who played this strategy perfectly. Say, maybe assuming the least one could double for less is $1 with a $25 original bet, so you could get that 4% in lol.
Or even, if much easier, just assume someone made a full double on all BS hands. I was thinking under this scenario the gain might be around 0.1% but maybe it's a heck of alot more when one includes the "doubling for less" strategy.
I also wonder what the gain would be to someone who only fully doubled his split aces only according to "normal" BS. From the casino's point of view, I wonder why they made this rule and what they might think it costs them.
Unfortunately, I'm always asking questions that I don't know the answer to
What do u use to figure this stuff out anyway?
SilentBob420BMFJ
Banned
in that situation, doubling for 1 cent would also be better than standing.. which makes me wonder, can you double for $1 on like a $2000 bet, i would assume so, which proves that doubling for less is never a good idea if its on your own hand, otherwise the casino wouldnt allow it (most likely)..k_c said:
im aware i created this thread, but i was really confused at the time after reading an article on scavenger tactics, and i was so confused that i actually asked a question about something i already have! i have a doubling/splitting for less chart (you have to read this article to understand (Dead link: http://www.blackjackforumonline.com/...vengerbjfo.pdf)).. this thread has turned into a doubling/splitting for less in general thread, which im glad.. i wish i could delete this thread or change the topic name, because i look stupid with my name under "where can i find a double down for less strategy"
you misunderstood what i said by so much, that i dont even know where to start.. i wont waste my time on it, but nothing you said is based on what i actually believe, you just didnt understand me.. how about instead of looking for me to say something stupid and for you to make fun of it, you actually read it carefully..Kasi said:
use at your own risk indeed.. your EV will go down, and your variance will go way up.. i feel like such an idiot for asking for a double down for less strategy, because i was confused at the time and i was thinking about scavenger play, cuz obviously doubling for less on your own hand is never a good idea, because double down plays ALWAYS have a better chance of winning than losing (notice i didnt say better than 50% chance of winning, so dont pick my words apart)miplet said: