Illustrious 18 - 16 vs. 10
You are really splitting hairs here. It is a very, very close play.
Quote: AxiomOfChoiceBasic strategy is an average, not what you do at count 0. The gains and losses as the count moves are not necessarily linear -- it's entirely possible that the median and mean are different (count 0 would be the median, basic strategy is the mean).
You are really splitting hairs here. It is a very, very close play.
9,7 vs. T, RC = -1
T,6 vs. T, RC = -1
In both cases, the TC is < 0, hence hit. So, basic strategy is consistent with stand with TC of 0.
Quote: teliot9,7 vs. T, RC = -1
T,6 vs. T, RC = -1
In both cases, the TC is < 0, hence hit. So, basic strategy is consistent with stand with TC of 0.
This is what I was originally going to post, but BS is to also cover when the player has more than 2 cards. These 2 cases are probably the most common occurrences of ending up with a 16 and outweigh the 2, 2, 4, 4, 4 hands of 16 where you would stay instead of hit.
Quote: teliot9,7 vs. T, RC = -1
T,6 vs. T, RC = -1
In both cases, the TC is < 0, hence hit. So, basic strategy is consistent with stand with TC of 0.
Also a good point, but I don't think that it invalidates what I said.
Indices are rounded to the nearest TC, anyway. And, besides, single-parameter counts are terrible for playing decisions -- you are taking a number that's barely correlated with the decision and rounding it off. And then worrying about a play that is so close that it barely matters. You have already thrown away too much precision to care.
Quote: AxiomOfChoiceIndices are rounded to the nearest TC, anyway.
I've always read that the TC is rounded down for index plays, so -1/49 would be an index of -1.
Quote: wudgedI've always read that the TC is rounded down for index plays, so -1/49 would be an index of -1.
What I meant was, the indices were rounded when they were created. They are not all whole numbers by some strange coincidence. The point at which it becomes profitable (on average) to change your play is not exactly the index number.
Quote: kw2107Quick thing I found, not sure how accurate it is... From my simulation using 2-deck, DAS, Hit soft 17, no surrender, 65% penetration, 1-8 spread, and the first 3 indices (surrender, 16v10, 15v10) I noticed an increase in player edge from 0.8% to 1.0%.
You mean insurance (not surrender), right? Without simulating it myself, I'd say that it looks about right.
Note that the indices are listed in decreasing order of importance, and you can round them and get almost all of the value from them.
Also note that you can find better double deck rules than that. I try to avoid H17.
Where can you find S17?
Quote: kw2107I see on the Vegas Blackjack Survey there are a few places. Unfortunately the Min gets bumped up beyond my limit...
I believe that in a previous thread, someone (I don't remember who) said that you could find a game like this downtown at one casino (I don't remember which one) for $25 minimums (or maybe he said $50?). I think that it was a thread talking about single-deck at the El Cortez, if you want to look for it. Or maybe that person is reading this thread and will repeat the information :)
Various MGM/Mirage casinos offer this game. It's usually black-chip (or higher depending on the night) but I have seen it with $50 minimums before. DD, S17, DOA, DAS, re-split to 4 hands. No RSA or surrender. A good game. You do not need a very large spread to beat it.
Quote: kw2107If I'm reading the Illustrious 18 chart correctly, it says to STAND when you have a 16 vs. a dealer 10 with a count of 0 or more. So with this logic, why is the basic strategy play to HIT on a 16 vs. 10? It would make sense to me if the count index was 1 or more... But with 0, that would mean you'd have a balanced deck, right? What am I missing here?
Four Points.
1) The running count is negative when you have a 16 v. 10 in basic strategy, because it assumes no other cards were dealt besides your two cards and the dealer's up card. Can you find a combination that leads to a zero count? No. If you extend the decks to infinity, then hit becomes better at 0. Fortunately, you don't play with infinite decks.
2) Standing at a running count of +1 instead of 0 will cost you about one minimum bet over the course of your lifetime, EV-wise. So it doesn't matter if you hit or stand on exactly 0, so long as you are using 0 as the basis of your index.
3) Zero count relates to deck composition. All basic strategy plays are based on an approximately zero true count, but not precisely a zero running count. For a new player, there's nothing wrong with relating zero count to basic strategy, but it is an oversimplification. If you do this, just keep in mind it is a fiction, if you decide to delve deeper.
4) Basic strategy does not cover multi-card situations. It only deals with the first two cards. Fortunately, there is only one exception to basic strategy--to stand on a multi-card 16 v. 10. Basic strategy also does not cover composition dependency; it averages the likelihood of all possible player hand compositions.
Quote: AxiomOfChoiceI believe that in a previous thread, someone (I don't remember who) said that you could find a game like this downtown at one casino (I don't remember which one) for $25 minimums (or maybe he said $50?). I think that it was a thread talking about single-deck at the El Cortez, if you want to look for it. Or maybe that person is reading this thread and will repeat the information :)
Various MGM/Mirage casinos offer this game. It's usually black-chip (or higher depending on the night) but I have seen it with $50 minimums before. DD, S17, DOA, DAS, re-split to 4 hands. No RSA or surrender. A good game. You do not need a very large spread to beat it.
The Mirage has it for $25 minimum. They may raise it on crowded evenings, but I was there last June and they never raised it in the three mid-week days I was there. You'll notice that the survey shows that MGM has it for $100 minimum, but I played it last February for $50 in the late afternoon. It was $100 in the evenings though.
Quote: geoffMGM Grand has $25 S17. This was only 2-3 weeks ago too.
The $25 S17 at MGM is on their 6 deck game. I was referring to the 2 deck game.
Now the third tell, that my pit friend says they look for is 16 vs 10 and whether the player plays it differently at different times. 16 vs 10 is problematic for several reasons. First, player 16 vs dealer 10 is the single most common hand, it occurs more frequently than any other hand, meaning there is more opportunity to see how you play it. Second, because the index for playing the hand differently falls right smack on zero, there are going to be many times you should play it each way, which will make for an obvious tell.
The way around this is something called 'counters basic strategy'. With counters basic strategy, you always stand on 16 vs 10, thus eliminating the tell. Now what are you giving up by doing so? Well, under normal circumstances, it is such a close play that you are giving up very little. But as a counter the times that the count is positive and you have bigger bets out, you are playing the hand correctly. So you are only giving up a tiny little bit when the count is neutral or negative. And if you exit aggressively on negative counts (wong out) as I do and as a counter should if possible, you are giving up even less. It works out to like 80-85% of the time you are making this play correctly, including 100% of the time when anything larger than your minimum wager is out.
I am not a big fan of cover plays. I believe a card counter's margin is too thin to go giving much back in cover, but this is one I make an exception on. With 16 vs 10 being one of the three top ways of identifying a counter, eliminating this tell completely is well worth the tiny cost, especially if you have any kind of longevity concerns as I do.
Quote: AxiomOfChoiceGreat post kewlj. But wouldn't 13 v 10 be the most common hand?
I don't know much but isn't a 20 the most common hand to be dealt, approximately 9% of the time? Closely followed by a 13 at about 8% of the time?
Am I confusing just regular players hands with hands of player v dealer?
Quote: AxiomOfChoiceGreat post kewlj. But wouldn't 13 v 10 be the most common hand?
16 vs 10 is not the most common hand. That's not really what I meant to say, but I did say it....lol, so my bad.
I meant to indicate that of the Illustrious 18 plays, 16 vs 10 is the most common hand, so if you are playing that hand correctly at all times, with the index for standing/hitting exactly on the zero count, there will be a pretty good sample size to evaluate. That's what you are trying to avoid. I am sure you know this, Axiom, but it was a good catch, so we can clarify.
Quote: kewlj16 vs 10 is not the most common hand. That's not really what I meant to say, but I did say it....lol, so my bad.
I meant to indicate that of the Illustrious 18 plays, 16 vs 10 is the most common hand, so if you are playing that hand correctly at all times, with the index for standing/hitting exactly on the zero count, there will be a pretty good sample size to evaluate. That's what you are trying to avoid. I am sure you know this, Axiom, but it was a good catch, so we can clarify.
lol, well apparently I got it wrong too.
But the idea of playing all hands in the way that is correct for high counts, but wrong according to basic strategy, is a good one. Ian Andersen talks about it in his book (although he takes it a lot further than I would)
Also not true. The most common is 15 vs. T.Quote: kewljI meant to indicate that of the Illustrious 18 plays, 16 vs 10 is the most common hand
Quote: teliotAlso not true. The most common is 15 vs. T.
My reference point was table 5.1 in Don's Blackjack attack 3, which lists the frequency of 16 vs 10 and 15 vs 10 the same, but of course 15 vs 10 being much less critical because the higher index means the variation will occur much less often. That's what my thought is about, avoiding this critical tell. I didn't bother to check Don's numbers, I'll leave that to you. You can take it up with him.
Quote: geoffYep. Since 88 is a split you lose one way to make 16 otherwise they would have the same probability.
Fair enough, Geoff. It doesn't change the point I was making.
Here, let me re-word the thought. Since 16 vs 10, is a very common hand and the strike point for strategy change fall right on zero, it provides a pretty easy read, with for anyone who may be evaluating your game. THAT is the point I was trying to make and what I am trying to avoid.
Even if you counted 8,8, there would be (in six decks) 3156 ways of making a 16 and 3456 ways of making a 15. More 15's.Quote: geoffYep. Since 88 is a split you lose one way to make 16 otherwise they would have the same probability.
Quote: IbeatyouracesWhat KJ is trying to say, is that 16 vs 10 is the most common hand where you vary your play. It's more rare to get to the point to stand on 15 vs 10 than 16 is.
Thank you, Ibeatyouraces.
I am currently training my second partner in as many years. Sometimes I try to make a point and am met with a blank stare and have to re-group and circle around and try to come at it from a different angle. I have accepted that conveying my thoughts clearly is not one of my strengths.
In this case, I feel like I made my point clearly enough, but there is some 'nit-picking' going on about minor details, that really do not effect the point I was trying to make. That is a little frustrating, but par for the course. lol
Quote: kewljFair enough, Geoff. It doesn't change the point I was making.
Here, let me re-word the thought. Since 16 vs 10, is a very common hand and the strike point for strategy change fall right on zero, it provides a pretty easy read, with for anyone who may be evaluating your game. THAT is the point I was trying to make and what I am trying to avoid.
I quite agree with you on that point and it is definitely something to consider. Personally I think it just depends on what type of game you play and how you act when you play. If you come across as an impulsive gambler, always fiddling with your chips and randomly increasing/decreasing your bet, then going where 16 says on the count can worthwhile. Always standing is good in general though and is sort of like how BS is to double A,8 V 6 even though the index is 1 because the value of doubling increases exponentially rather than linearly.
Teliot, if you assume infinite deck are they the same probability? I don't usually do the actual math and just figured based on the number of combinations each could make, but since getting an 8 makes the next 8 less likely I can see how 15 is more prevalent in live games.
Quote: IbeatyouracesAlso, in a 4 or 6 deck, S17 game, you stand on a four or more card A,7(soft 18) vs an ace.
Didn't know that. The usefulness of this knowledge is profoundly infinitesimal. Doesn't really fit, but I'd plug it into composition dependence, which is never a consideration. I would never consider the 16 v. 10 exception, but for surrender. Like Pluto, I don't want this to be an exception so I'm calling it something else.
