WHAT TO DISCARD?

Ask questions, discuss and debate your strategies, euchre polls and more
Post Reply
irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

WHAT TO DISCARD?

Unread post by irishwolf » Thu Feb 06, 2020 5:14 pm

The QS is the up card and your side is down 7 to 3. As the dealer you decide to go alone and hold JS 10S AD KD AC. What card will you decide to discard and why did you choose this card?



RedDuke
Posts: 473
Joined: Sun Dec 02, 2018 2:22 am

Unread post by RedDuke » Fri Feb 07, 2020 10:51 am

The Ace of Clubs. By discarding that, you have a two-suited hand. It's one fewer spot that the opponent can trump in. For instance, if first seat leads the King of Clubs and his partner trumps it, your Ace of Clubs would be pretty much worthless even though it is a boss card. If you don't have that Ace though then you can easily overtrump his partner.

With that said, I can't really fault someone for discarding the king and keeping both Aces, but it's not what I'd do.

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Fri Feb 07, 2020 1:52 pm

Ok Richard, what does your BPS say about the point evaluation of an AC void vs not discarding the AC?

Richardb02
Posts: 748
Joined: Mon Dec 10, 2018 8:57 pm
Location: Florida

Unread post by Richardb02 » Fri Feb 07, 2020 7:20 pm

Up (Card_Q-S) Score 3-7, Your S4 hand:
(Card_J-S) (Card_10-S) (Card_A-D) (Card_K-D) (Card_A-C)
If Ac Discarded:
0.50 R1 S4
1.00 Js
0.50 Qs
0.25 Ts
0.50 3 Trump
0.50 Ad
0.75 2 Voids
4.00 Points (Not in BPS but I would consider +.25 for Kd)

Ac Held, Kd Discarded:
0.50 R1 S4
1.00 Js
0.50 Qs
0.25 Ts
0.50 3 Trump
0.50 Ad
0.50 Ac
0.25 1 Void
4.00 Points

So BPS calls the hands equal. But as I have pointed out before:
1. BPS is most effective with average hands, 1.75 to 3.75 BPS points or 11/2 to 3 tricks. This hand is stronger than 3 tricks, so additional analysis Is required. IMO, it is mandatory with a hand this strong!
2. The Kd is a Boss being paired with the Ad in this hand. Additionally because I have 3 trump and can pull trump, its value is even stronger. As noted I would definitely consider adding 0.25 point in general, even though it is not covered in BPS. And because of the 3 trump, including a Bower, I would consider the Kd equal to an ace, 0.50 total points. So using BPS(plus, factors rarely important enough to consider), this hand is worth 4.50 points, I would discard the Ac.
3. I use BPS in real game situations. So it is simplified to some degree. If I analyze with BPS and it is close, I will make a “gut” decision. My gut decision with this hand is actually a vivid memory of an OE lesson that taught that creating 2 voids in a similar situation is better than 2 aces. My mindset with BPS is not to justify BPS but to continue to improve my game and by doing so to improve BPS. As I covered in #2, I can use visible factors to continue analyzing and add quantified values that are beyond the normal BPS and they make sense. So BPS is not so much a perfected system as a method of analyzing a hand, first for ordering/calling but even extending to guidance about how to play the hand. If BPS says I have “power” in my hand, I play power. If BPS says I am short on power, I play a “finesse” style

IMO, my analysis matches RedDuke’s analysis. It matches OE lessons. The value that BPS adds, is assigning a quantitative value to each of these “visible factors.” Finally, the BPS concept includes, if it is a close decision (within 0.25 point), base your decision on your storehouse of knowledge and experience.

IMX (in my experience), I have enjoyed BPS Basic to gain confidence as I learned the game. (I started playing October 2017). It was much more enjoyable than the floundering I experienced with weakly defined factors as opposed to numerically defined factors. BPS readily expanded to cover all of the OE lessons. BPS continued to expand to help me understand posts on the forum (OEF for OE Forum). BPS allows me to easily translate from BPS values to readily described scenarios, ie 3 near certain tricks. So I can participate with the much more experienced posters on OEF. I definitely appreciate all y’all (I live in Florida) allowing me to participate on the OEF. Now, here in February 2020, I can play decently on Monday nights with “Euchre Legends.” The feedback I receive from the analysis of those hands, provides me even greater enjoyment playing the game of Euchre. Thank you OE & OEF.

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Fri Feb 07, 2020 8:31 pm

Great write-up Richard. I was only interested how your system would measure the two hands. A little surprised equal but you explained that.

I agree also, the void is more important in the hand above as the AC is double jeopardy of the risk having two suits to get trumped by S3.

What is little different about the hand is that there are three trump cards if guarded can also stop this loner so does that may holding two off suit aces worthy? I firmly believe NO! The euchre rate is low, very low, but still exists with certain hand combinations.

~Irishwolf

Richardb02
Posts: 748
Joined: Mon Dec 10, 2018 8:57 pm
Location: Florida

Unread post by Richardb02 » Fri Feb 07, 2020 10:51 pm

The equality is that an Ace=.50 and the difference between 1 Void and 2 Voids is 0.50. They are roughly equivalent but that is an apples and oranges comparison. Apples being average (acidity) and oranges being strong (acidity), to enhance the comparison. The strong hand (oranges) makes the AKd preferable over the average hand Ad Ac.

If I was ordering with a weaker hand, BPS <3.75 (apples) or not quite 3 strong tricks then I would keep both aces. Why would I order? I’m behind 3-7. BPS will add up to 0.75 points to compensate for being behind by 4 points or more. So if my hand is only worth 3.00 points but the situation adds 0.75 points for a 3.75 value, we have to analyze both evaluations. We order alone but we realize that we only are confident of 2.4 tricks. So we hedge being euchred by discarding the Kd and keeping 2 ace bosses. BPS helps us determine what to discard as well as ordering and playing the hand!

This hand is so strong, greater than 95% chance of earning 3 tricks, that it is mandatory to stop considering the possibilities of being euchred and instead focus on maximizing your opportunity to earn 4 points and tie the score. That is the key advantage of keeping AKd and discarding Ac.

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Sat Feb 08, 2020 9:32 am

I would evaluate the hand (JS QS 10S AD KD) as 4.0 tricks and making a loner on the average around 40% and euchres almost nil. Even an opponent with JC AS KS or 9S will have difficulty euchring me as long as I do not lead trump for the second time until trick 5.

Richardb02
Posts: 748
Joined: Mon Dec 10, 2018 8:57 pm
Location: Florida

Unread post by Richardb02 » Sat Feb 08, 2020 9:47 pm

irishwolf wrote:
Sat Feb 08, 2020 9:32 am
I would evaluate the hand (JS QS 10S AD KD) as 4.0 tricks and making a loner on the average around 40% and euchres almost nil. Even an opponent with JC AS KS or 9S will have difficulty euchring me as long as I do not lead trump for the second time until trick 5.

IrishWolf, let’s review where we disagree. BPS is 1.25 points per tricks. So even at 4.50 points, the 95% certainty is 3.6 tricks, not 4.00. Do you see something for which I am not accounting?

From our parallel post: The probability of taking 5 tricks is rather weak.
There are 3 trump that if protected or available because of a void (Irish’s scenario) then my short cut to “guesstimate” probability is:
3 trump, 2 parlays, each parlay equivalent to 1/2 of a stopper.
So 3 stopper equivalents.
Each stopper has a 55.5% probability of being in an opponents hand.
That leaves a 45% chance of Loner success for 1 stopper equivalent.
45% x 45% = 20% for 2 stoppers and about
20% x 45% or 10% for 3.

Now, by memory, there is only about a 22% of a void. So without going into deep calcs, let me use 0.78^6= 22% probability of getting 4 points. 40% possibility for 4 points just doesn't seem reasonable. How do you arrive at 40%?

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Sun Feb 09, 2020 11:45 am

Richard said,
"Now, by memory, there is only about a 22% of a void. So without going into deep calcs, let me use 0.78^6= 22% probability of getting 4 points. 40% possibility for 4 points just doesn't seem reasonable. How do you arrive at 40%?"

I don’t always like to divulge proprietary information as I am writing a book on euchre. I will do it this time. I did a test for my own information on successful loners of discarding the AC vs KD. I did 300 hands in sets of 75 hands each for reasons of looking at variability as well. These numbers are actually a little bit low for successful loners as I will why. I had S1 always leading to the Club void or the Diamond void so S3 could trump as a stopper. If S3 had the JC guarded but not when the KS or AS was guarded (as that is what I would normally do), he did not trump with JC guarded as that is how it would be normally played not knowing the strength of the dealer. If the KS AS or JC was guarded, of course the loner was also stopped. And I did not continue the hand as the purpose was for successful loners based on the AC discard vs the KD discard (and for time).
The results for loners overall (300 hands) was 27% loners for discarding the KD and 34% successful loners when discarding the AC. The sets for the four varied greatly for each, 26%, 28%, 38% and 44% (34% average) for discarding the AC. I was surprised at the spread, but it is what is as euchre has a high variance. I dealt these cards in 1s not 2s and 3s to achieve more randomness in the cards (so it could be worst, lol). There are ways to check for randomness which I did like for four trumps S1 or any player will be void 23.4%. This varied but was close.
So why are these numbers a bit low? S1 has a choice of three suits to lead but I was favored leading to S3 void first AC then the Diamond playing same hands both ways. The reason is I wanted worst case as my study was on the discarding. However, in reality the lead might have been to lead a heart or the diamond instead to the AC void. Thus, I think these numbers might be (guessing) as much as 3 to 5% higher for loners with S1 not leading deliberately to the void. For euchres, these were nil (4 total) as S1 and S3 has to have two trumps each as well as leading to the correct voids. Even for having three trumps hard to achieve get a euchre. As for having four trumps and I had none for 300 hands (I think the number is about 1.6 in 1000).
I am going to do more hands to get this to 400 for 95% confidence level when time permits. As to how it compares with your calculation that’s for you to figure out.
But I think your issue is not just with the voids occurring but also when S1 or S3 has a singleton trump, each, or in combination with a void.

~Irishwolf

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Sun Feb 09, 2020 12:03 pm

Richard,
I think you should revisit your calculations. There are four ways to achieve a loner with the hand above, revisited this:
S1/S2 has no trumps, each has 1 trump (47%). S1 has a void/ S3 1 trump, and S1 has 1 trump and S3 has a void (x2). This add up to 49.5% minus the leading to the AC or Diamond void of S3 - 23.4% + a trump. So that going to be 49.5 - ~17 = 32.5% right at my actual test results. Maybe I don't need to bother with another 100 hands, lol.

Richardb02
Posts: 748
Joined: Mon Dec 10, 2018 8:57 pm
Location: Florida

Unread post by Richardb02 » Thu Feb 13, 2020 7:21 pm

Thank you IrishWolf for sharing your proprietary information. Congratulations on completing 300 simulations of this hand. Congratulations in advance for your soon to be published euchre book. Good luck on a profitable run. Count me in to buy a book.

Your guidance has lead me to a simple approximation.
If a Singleton can stop 4 points then there is a 45% (8/18) chance
If 2 Singletons can stop 4 points then .45^1.75= 25%
If 3 Singletons can stop 4 points then .45^2.25=17%
If 4 Singletons can stop 4 points then .45^2.50=14%
I reduced the power by 0.25 each time.
I did not reduce the power previously.
If a protected Trump can stop 4 points then increment by .3
If a trump +Void can stop 4 points then increment by ,2
I see Limited benefit in reducing the power with multiple parlays.
The increment is too small. Keep it simple, when reasonable, IMO.
Our hand had:
0 Singleton Stoppers
3 Protected Trump stoppers = 0.90
3 Trump + Void Stoppers = 0.60
So raise .45 to a 1.5 power = 30%

I like the EV down to a 10-12% expectation of 4 points.
Yes, it is fuzzy math, but it passes my smell test.
It reflects my positive experience of going Alone with a weak Loner hand that is weaker than weak (usually a Hail Mary).
It is, at a minimum, worthy of testing.
All it requires is 3 near certain tricks. There is no need for BPS.

Post Reply