"Surrender Trap"

Here is my “old” article on the surrender trap that I made available for a couple of weeks, over a year ago on Ken’s site. I wrote it keeping in mind an average reader, providing some numbers but trying to make it accessible.
I hope it inspired James and Previn, but I wouldn’t be surprised if they applied their talents without any additional motivation. Their work is absolutely top notch and in depth.
The only thing I would add to a fuller version is a “half trap”-a bet that could be used in specific situations against experienced players, where the only chance to get the low for BR2 is the minimum bet, but both players losing their hands causes a tie.

S. Yama

 

Surrender Trap

“Surrender Trap”

Tournaments that offer surrender present skillful players with a possibility of using a special technique increasing chance of qualifying to next round or winning it.
Many players realize that “taking low”, which is relying on your opponent losing regardless of what happens to their own hand (chance of ~ 48%), is better than chance of winning their own hand (44%) or winning a forced double bet (31%).

If you are playing in a tournament that offers surrender and you have only one important competitor and you have lead of at least five minimum bets- you may try this:
Bet your lead plus about twice the minimum bet, preferable a rounded amount that looks like uncalculated, spontaneous bet.
For example, you have 10,900, your opponent has 9,200 and limits are minimum 100 maximum 10,000. Your lead is 1,700 - bet at least 1,900 -better yet, an even two thousands.
Your opponent in the fervor of the tournament play may forget about surrender and calculate that if you lose your hand he ends up with 300 chips more that you would have if he bets only minimum (regardless the outcome of his hand). His bet of 100 or 200 –even if doubled won’t let him overcome you if you surrender.
Your losing 1,000 on surrender and his winning doubled, or even tripled bet (600) is not enough to overcome 1,700 gap.
Try not to bet more than 120% of your lead. If you bet more than your lead plus 20%, your surrender loses more than .6 of the lead and opponent winning doubled bet the size of your “overbet” (.2) wins .4 of the lead overcoming your lead. Weaker players often make mistake of betting the whole “overbet”.
(X is the “overbet” and Y is the lead. 2xX+1/2Y+1/2X=Y 4X+Y+X=2Y Y=5X)

The scheme works better when you “overbet” just minimally, two to four minimum bets.

Even very experienced tournament players forget about surrender option. Don’t remind opponents at the table of this possibility; don’t surrender, otherwise surrendable hands, if the size of your bets is not significant.

Also watch out, especially if cards are dealt face up, for the possibility of your opponent winning triple bet and ending up with more money than you would have after surrendering your bet. In such a case, you may be better off by taking a chance of pushing or winning your hand.

Be careful not to make this bet in tournaments with limited max bets where your lead is slightly more than one third of max bet – bet two-thirds of max bet. If your lead is more than half max bet -bet a chip less than your lead. Bet max if you are quite confident that your opponent will match your bet with max bet, as well.


S. Yama

 

What about surrendering in the middle of a tourney?

What advice do you have on the more general use of surrender in a tournament? I tend to use it more than BS dictates (as most people seem to) but it's based on a general understanding that in certain situations I maybe a dog compared to a nearby competitor and so I surrender to minimise my potential loss against them. I don't know the maths of it and whether I'm doing the right thing. Can you offer any advice on good situations to surrender on hands other than the final hand?

Many thanks in advance.

Reachy

 

Nice article S. Yama. Thanks for posting. Good stuff.

 

Reachy,

In most cases there is no need to surrender more than the basic strategy says. Learn and stick with the basic strategy for most of the hands played. A proper surrender saves you relatively very little in terms of EV, and does even less for your tournament chances.

Actually, in most cases do not surrender in order to not remind other players about this option.
It is unlikely that an inexperienced player would make a mistake in final hands and surrendered unnecessary.
There is, though, a better chance that your keeping surrendering during the round may help her or him “come up” with the surrender that just can happen to be the best play.

Even during the end play, and with bigger bets and with though to win hand, benefits of surrender rarely exceed total benefits of winnings (doubling may be the best option) and losing.

Look up for situations where you are relatively close to the leader who acts after you and has a difficult hand to surrender, like total ten, eleven, or pats: eighteen, nineteen, even twenty if the chance of the dealer pulling twenty-one is better than you winning a double bet that would get you ahead of the leader, or the leader played in front of you and stood on his pat/stiff hand.

Most of the time surrender is helpful when you made a sizable bet that would advance you if you won the hand but other players took the low on you. If your hand was not that good, especially when the dealer showed a strong card, surrender might regain you the low.

Yet, another type of play is when you tried to catch up the leaders and made a big bet but got a high stiff vs. the dealer Ten or Ace, you might consider surrendering if winning a double maximum bet in the next hand would give you a chance to beat your opponents but losing the hand would take you out.

Be careful with surrender plays, they are powerful but easily misplayed.

S. Yama

 

Thanks S. Yama, very interesting useful information. I follow BS pretty much by the book until the last few hands depending on my BR and postion on the last hand. Because I am not yet versed in the statistics of BJ, in most potential non-BS situations I have to make inferences about the likely relative probabilities based on how much of a deviation from BS I think I am making. For example I surrendered a 14 against a 10 last night figuring that if BS says that 15 and 16 should be surrendered then 14 can't be that much better (or should it be worse?) I also had about a 1/2 max bet out trying to catch BR1 who had a pat hand if I remember rightly (18?). What do you think?

 

Doesn't matter how close to the basic strategy it is.

Reachy, you should pay less attention “to likely relative probabilities based on how much of a deviation from BS” you make. It is disproportional to the effectiveness of your play.

For example, you need to beat one opponent. Let’s assume for the clarity of the situation that he will make minimum bets no-matter-what, and he will be betting in front of you, two hands left. Let’s say, you have 1,000 and he has 1,450. Bets are min. 5, max 1,000. So, you bet 500.

The dealer is showing a deuce, and you have a stiff (it makes no difference if it is twelve or sixteen).
Surrendering your hand to the dealer showing a deuce is not even near BS play, yet, it is the best tournament play. You lose your hand if the dealer makes a hand. If you lose this hand you are out of the tourney, even if you get a bj in the last round with the all-in bet.
Surrendering this hand gives you an option to bet all-in in the last round and overcome the leader. The chance of winning a random hand (that will be the last round hand) is better than the chance of the dealer breaking with the up-card two.

It is not how close the surrender play is to the basic strategy play, but how much you improve your chances to achieve the goal.

Now, going back to your case of surrendering 14 with about half-maximum bet out, trying to catch-up the leader near the end of the tournament.
Tell us, your estimate (it doesn’t have to be exact -it's not the point here, just some numbers so we can play with them), or simply, what you think were the chances of your final winning/advancing, for four different scenarios: a) you lost that bet, b) you won that bet, c) you doubled and won that bet, d) you lost half the bet by surrendering.

S. Y.

 

Oh my God! I'm out of my depth....

This is where I demonstrate my complete lack of experience and knowledge in matters of BJ!! I've prepared my dunces hat and will be standing in the corner of the room facing the wall shortly !

Firstly I'm not sure I understand you question exactly. I think you are asking me what I think my chances are of winning the table in either of the 4 outcomes specified rather than the specific chances of those circumstances arising. Am I right?

Anyway I have put the hat on now so here goes. It's a shame I can't remember the specific BRs and bets but for the sake of this question I'll assume that

BR2 (Me) - $1000, bet $250

BR1 - $1300, bet $250

This is not the final hand but maybe a couple hands away from it. I acted first. I probably should have bet more, say 300+ but that is not the issue here.

Cards are:

Me - 14

BR1 - 18

Deal - 10 (?)

This is where the holes probably start to appear. My assumption was that the most likely dealer outcome was 20. In that case we would both lose so but if I surrendered I would close the gap on BR1 by $125. Anyway to your question.

These are my estimations (and I will qualify them as barely educated guesses) as to what my chances of winning the table would be if:

a) I lost my bet - 15% because I would probably have to win a max bet DD (or get a BJ) on the final hand to have any chance. BR1 could cover it though with a single bet of $455 for the high and low. If BR1 wins - 0%; if BR1 pushes - 10%

b) I won my bet - If BR1 loses - 80%; If BR1 wins - 30%; if BR1 pushes - 50%

c) I won a DD - If BR1 loses - 90%; if BR1 wins - 50%; if BR1 pushes - 60%

d) I surrendered - If BR1 loses - 60%; if BR1 wins - 10%; if BR1 pushes - 20%

Can I just re-iterate that no mathematical skill or charts have gone into these figures. I wouldn't even know where to start!!!

I'm in the corner now. I'm facing the wall....

Cheers

Reachy

 

Chances

You’ve got it right.
It takes a lot of practice to be able to precisely estimate chances, but the only objective is to find a better option at the time of decision. You will have time to find out by how much later.
You may not know if winning a hand gives you, for example, 40% or 30% chance for the final success. For practical reasons you may skip worrying about pushes, pushes (yours and your opponents') will work out somewhere in the middle. And losing could still be worth more than 10%. If you assume that you win and lose close to fifty-fifty , your total chance would be about twenty five percent.
Now all you need to do is to figure out what are the chances if you surrender. If it is more than 25% then you surrender, if it is less then you don’t surrender.

The specific numbers sound scary but if you work with them then you develop a “feel” -that is not just a shot in the dark but a truly educated guess about (cor)relations.
This way would also teach you to assess a lot of other situations where you need to decide if it is more beneficial to deviate from basic strategy to go for a swing, or to increase chances to correlate with your opponent(s), etc..

Going back to you case.
You further separated the four possible results I asked you about by possible results of your opponent’s hand. Great – that’s what it is all about.
Now we have twelve possibilities. That needs to be reduced to a manageable level.
If you include the necessary correlation, using your numbers (they don’t have to be absolutely right, but at the time of the decision, the best you can manage) gives you final success about 25% if you hit to seventeen, only 16% if you double down, and 37% if you surrender.
Surrender was the best play by far.
However, we may need to work on your numbers, lol

It is extremely important to know (as best you can) the chances of success based on position and number of hands left to play.
You wouldn’t know where to start? In the beginning, of course.
Well, actually, we could start with your example, three hands to go or at next-to-the-last hand. Let’s make it with specific brls. It bothers me a bit that you did or could have bet 250 when the gap to the leader was 300. So, take a pick, make the bet 260 or the gap 240. Also, specify BR1 bankroll and bet.

S. Yama

 

Thanks Mr Yama.

I think I need some time to digest what you have said before I make an informed response.

I'm not sure what you want me to do here. Do you want me to set up a scenario? If so I'll keep than BRs and bets as they are and make the gap 240.

Also do you think it is always necessary to bet to take the lead?

many thanks

Reachy

 

situation

Reachy, please excuse my pitiably contemptible English. I hope you don’t loathe it in a way only a genuine Englishman could pronounce the word loathe.

Also, I meant bet of 310 not 260. The point was that your bet should be meaningful.

On the side note, to answer your question, no, it is not necessary to try to always take the lead, often not even with a few hands left. We may get to this using your example.

I can try to show you how to estimate value of different decisions for the final win/advancement using your specific example.
Let’s make it 5 min, 500 max, surrender available, only one winner, BR1 has 1,230, you, as BR2 have 1,000 and bet first. Present BR1 will be betting first on the last hand. Your bet is 250.
Reachy, make up the BR1 bet as a response - it doesn't have to be a very good bet.
We will try to do it assuming that you’re very skilled player and your opponent slightly better than average.

S. Yama

 

Your English is far better than any attempt I can make at any language other than my own so I certainly don't loathe it. However I do like the word loathe and try to use it in conversation as much as possible

I'd also like to take this opportunity to thank you for taking your time to impart your knowledge to me (and whoever else is reading). I really do appreciate it.

Anyway to your question, my bet as BR1.

The scenario, 2nd to last hand, 1 player advance, BR2 acts first and we have to figure out he best way for BR2 to win the final hand.

BR2 - 1000 bet 250

BR1 - 1230 bet 150 (covers BR2 BJ and can cover DD with DD)

There you go.

Look forward to your next post.

Cheers

Reachy

 

Gotta catch the plane.
Be back Monday.

S. Y.

 

Reachy Avatar

Reachy, shooting for membership in the Droogs. You've got a lot of response to your posts, but I think you'd get even more if you'd include Emma with Stead.

 

Who's Stead??? That's a picture of me relaxing after a hard day in the Stock Exchange. In fact I've just finished a nice cup of tea and am about to tuck into a fish and chip supper!! Mmmmm!!

Anyway I was more of a New Avengers fan so it'd have to be Purdey for me.



Pip pip!

Reachy

 

Choices: Decisions/Decisions.

Purdey or Emma?

Marianne or Ginger?

I guess, if one is stranded on an island and has a choice, and can't have both, then one is better than the other.

But after escaping from the island, seeing Parie, or London, the choices become more difficult. Good bye Ginger and Marianne and darn Mrs Peale, hate to see you go, but Purdey!

 

hit, surrender, double down

Okay, there are so many aspects of what we were talking about that now I am not sure where to start.

With BR2’s bankroll of 1000 and bet of 250, BR1, out of his bank of 1230, should bet at least 185.
Bet of 115 is very good, and 150 or 155 even better.
Bet of 185 not only covers BR2’s bj and both players winning dd, but additionally, BR1’s blackjack locks out BR2 if he loses.
Note that it is convenient and can be fun to find those situation where dd or bj covers the other player’s dd or bj, but practical results, especially if it is not the last hand are very small.
In our specific case (bet of 115 versus 185) the difference in the final advancing would be around one percent (from 0.75% to about 2%, depending how well they play).

As to your estimation of the chances for different scenarios - you were very close in all of them. Generally, you slightly underestimated BR2’s chances.
Using your “feelings” chances for the next-to-the-last hand playing decisions (h17, dd, surr.) were 25%, 16%, and 37% respectively. While the specific case was more like: 38% for hitting, 21% for doubling, and 35% for surrendering.
The difference of three percent between hitting to seventeen and surrender is too small to really call one play better than the other, especially when the margin of uncertainty, created by unpredictability of individual players, could be bigger than that.

Also, this is a good example where counting could easily skew the right play. Note the double incentive: with more small cards left in the shoe hitting BR2’s fourteen and winning against the dealer Ten upcard increases, and with more big cards left it decreases, and the chance for the dealer ending up with twenty is more likely, so surrender becomes an interesting option.

I think I will start a new thread, as the direction of our discussion took us to subjects that have very little to do with the surrender trap. I will try to get more general and frame subjects more broadly, but we may return to this example.

S. Yama

 

Thank you.

It's often the case that one question leads to another. At the end of the day the question is always "what would my best play be?" and that is a complex question. That's why I love BJ.

Cheers

Reachy

 

Surrender BACK Trap and mid game surrender.

When you have a relatively late bet position on the last hand in a game where surrender is allowed, you can do what I call a surrender BACK trap.

If you are BR2 or 3, and bet after them, often they will bet max or close to it. this often puts you in a forced double position to take the front (high). If you take the back (low) by a min bet, you're exposed to a leaders surrender re-taking the back. This is where I set the surrender back trap. I quickly figure what BR1's back will be if he/she surrenders, and if my current bank roll exceeds that amount, I bet 2x my lead over their surrender back less 2 min bets (the "less" amount depends on how the house works with surrender, minimum units, and surrendering odd amounts). The Idea here is that you let BR1 see that their surrender can take the back away from you, but leaves you able to re-take it with your own surrender.

Here's a scenario to help illustrate.

max bet 500, min 10
Advance table 3 move on.
player 1 has 1500 and bet 10 :back 1490
player 2 has 515 and bet 500 :back 15
player 3 has 300 bet all in, 300 :back 0
Player 4 has 405

BR2 surrender back is 265. My lead over that is 140, doubled is 280, less 20 is 260.

Player 4 has 405 bet 260 :back 145

Player 2 had an average hand and surrendered. His 265 back looked good to him compared to my 145 back. At that point I had a lock to advance by springing the trap. My back, 145, plus my surrendered half bet back of 130 = 275

This Tool only works when you are in a situation with the following conditions.
1) you are within 1/2 max bet of the leader(s)
2) you bet after them
3) they bet big or max
4) your opponents are good but not great and haven't read this post or been burned by this move before.

The quick thinking goes like this...

As the others bet, you quickly estimate their surrender back, figure your lead over it, double that lead less a little to pick your bet size.


Next, mid game surrender:
Basic Strategy indicates surrendering in situations when your best choice between hitting or standing still produces an overall losing percentage of > 50% of your bet. For example, infinite deck model

player 17 (and stand) Vs dealer Ace
Win Push lose EV (Expected Value)
17 / A 0.2007 0.0830 0.7163 -0.5156

player 17 (and hit) Vs dealer Ace
Win Push lose EV
17 / A 0.1828 0.0551 0.7621 -0.5794

player 17 (and Surrender) Vs dealer Ace
Win Push lose EV
17 / A 0.0 0.0 0.5 -0.5000

About 1/2 of all 2 card hands are in the Stiff range (12-16),
and about 1/2 of those will be facing a dealer card that's good (8-Ace).

In the Mid game, when you've got fairly good sized catchup bet going, and you can't afford to waste too much of your Ammunition (Bankroll) on a stiff hand, I will surrender just a little more liberally than indicated by BS.

I went through all the outcomes and picked all hands where the EV was -0.4 or worse, and came up with the following situations where I might consider mid game surrender.

17, Vs dealer 8, 9, 10, Ace
16, Vs dealer 7, 8, 9, 10, Ace
15, Vs dealer 8, 9, 10, Ace
14, Vs dealer 9, 10, Ace
13, Vs dealer 10, Ace

The memory trick I use to remember this is "17 is trouble, and if your cards plus the dealer card = 23 or more (Ace = 11), give it up.

Remember, you have only 2 types of Ammunition, bank roll and hands left to play. If either one runs out before you are BR-advance, then you are BR-out.
Surrender preserves bankroll, but wastes a hand, so don't over use it.

PapaTreat (friend of Re-buy Bob's)
PapaTreat@yahoo.com

 

Excellent post PapaTreat!! Thanks.

Cheers

Reachy