First lead on alone variation

Ask questions, discuss and debate your strategies, euchre polls and more
Post Reply
Streblerm
Posts: 24
Joined: Fri Mar 22, 2019 10:58 am

First lead on alone variation

Unread post by Streblerm » Wed Aug 19, 2020 2:45 pm

Recently started a new online app that I like. Except, any time you call a loner the first lead goes to the left of the loaner caller instead of left of the dealer.

I have never heard of this variation and it has burned me more than a few times when I called a loner from 1st position expecting to lead.

Is this a common regional variation? In rated play it’s the only option for the app. I mentioned it to the developer and they responded as if I was crazy for asking the question.



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

Unread post by Richardb02 » Wed Aug 19, 2020 8:27 pm

You are talking about Karman Games. Wes introduced me to the sight several months ago. I find the positives on Karman outweigh the bad. I get a reasonably competitive game in less than 1 minute! The games against their bots are reasonable. (If you want really, really good games, you need to join us at WoCG, worldofcardgames.com, Mondays at 850 PM ET).

The “Loner plays from S4,” feature, is a corruption of the original game. However, it must account for some of the popularity of the sight! And, as you found out, they have no intention to change their variation. So our choices are to play private games, or to find another site, or to adjust your strategy.

Personally, I have not changed my strategy yet. There has not been any major impact on my expected results. (I can quantify my results easily, because I use my BPS plus Advanced BPS for Loners). Wes and I have discussed the “Go under” variation. We agreed that the adjustment would be 1/2 of a klick. That is 1/2 of the minimum adjustment when using BPS. Our conclusion therefore, is that only marginal hands are impacted. So, it is not worthy of a great deal of concern.

You raise what is probably the most critical negative situation with their rule. That is an S1 Loner. As you stated, you lose the opening lead. But in exchange, you now get to play from S4, and have the last play on trick 1. That is roughly a wash on a R1S1 Loner. Now R2S1 is more nuanced. The value of a Next Loner, as compared to Next, has been significantly reduceded. Personally, I make that adjustment, as part of my Advanced BPS, even playing the original rules. (Playing a R2S1 Loner, I don’t have the advantage of P being able to take a trick, because P is also strong in trump (80% of the time). I simply reduce the value of R2S1 Next to a minimum (0.25 points in BPS, or 1 click in more generic terms), any time that I am evaluating a Loner(traditional or Karman rules). In comparison, playing a R2S1 Next is worth 0.75 points or 3 klicks. Keep in mind, as Wes has taught me, a R2S1 Next also has defensive purposes, that is 1 klick. Partner, sitting out the Loner explains the logic, for the 2nd reduction.

I would appreciate more feedback. Please send me details on your hand, if available. Wes, please chime in. And, anyone else please post your thoughts. As a bonus, does anyone know what happens with a R1S1 Loner? Specifically, does S4 still get to pick up the Up Card and discard a card?

Streblerm
Posts: 24
Joined: Fri Mar 22, 2019 10:58 am

Unread post by Streblerm » Thu Aug 20, 2020 11:22 am

It is Karman. I do like the play and I'll probably continue. I'd been playing A-Star for a long time. No problem finding partners, clean gameplay, hoyle rules/stick the dealer. It might be paranoia but I never felt like the deals were really random, or maybe they are too random. Either way...Karman seems more like playing with actual cards. I also don't like "going under" but the advantages/disadvantages are a little more clear with that variation, as you pointed out.

I will admit, I don't currently use a scoring strategy. I'm not opposed to using one and I appreciate the analysis and quantification. When I say I got burned, I'm not talking about getting euchred, just not taking all the tricks. I probably wont change my loner calling strategy much, other than calling loners a little more aggressively from S2 and S3. I guess I just need to temper my expectation of the outcome sometimes.

As you suspected, I'm talking about R2S1 Next loner calls.I don't have a photographic memory but last time the cards went something like this:

(Card_A-H) was turned down
S1: (Card_J-D) (Card_A-D) (Card_Q-D) (Card_A-C) (Card_Q-C)
S2: (Card_J-H) (Card_K-C) (Card_K-S) (Card_Q-S) (Card_9-S)
S4: (Card_K-D) (Card_A-S) (Card_J-S) (Card_10-S) (Card_9-H)

S2 led (Card_K-C) S4 Trumped with (Card_K-D) and I was forced to follow suit. Had I played from the natural position, I would have pulled all the trump with the right and gotten the loaner.

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

Unread post by Richardb02 » Thu Aug 20, 2020 9:23 pm

My take has been for decades, that all games of chance are streaky. I have concluded that online or computer generated deals are just as streaky as real cards. It could be a matter of expectation. I expect streaky card distribution. You expect something different.

The BPS is most helpful with edge and near edge hands. It is also simple enough to let me make crisp analysis in real time, in real live games. I utilize another process to determine my expectation of taking 4 points.

Using that process, I estimate that, based on the hand you posted, you had a 75% chance to take all 5 tricks from S4 and an 87% chance to take all 5 tricks from S1.

That isn’t a large difference. So I suggest, that you simply got hammered by unfortunate card distribution, much more than the S4 variation.

I feel your pain, because I have experienced the same disappointment! I have had to learn to chalk up missing the Loner to card distribution, let out a deep sigh of disappointment and get focused on the next hand.

Streblerm
Posts: 24
Joined: Fri Mar 22, 2019 10:58 am

Unread post by Streblerm » Fri Aug 21, 2020 10:54 am

I get what you’re saying. When you break it down statistically, the S4 loaner isn’t as big of a deal as it seems. It’s a bitter pill to swallow when that loaner would have walked from S1. I still would have called it from S4,but with an adjusted expectation.
In the end, the error was mine in forgetting that I wasn’t leading from S1.
I was also curious how common this variation was. I had never even heard of it. I know add-on rules tend to be regional, so I want to know who to blame for this nonsense.

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

Unread post by RedDuke » Fri Aug 21, 2020 6:42 pm

My take has been for decades, that all games of chance are streaky. I have concluded that online or computer generated deals are just as streaky as real cards. It could be a matter of expectation. I expect streaky card distribution. You expect something different.
I too wonder how random the KARMAN app is. Over the past 100 games, I've seen some very strange things that should be so statistically unlikely as to be basically impossible.

As dealer, the turn card is my only trump 61.5% of the time. Statistically, the dealer should have at least two trump (including the turn card) 70% of the time.

On 41.6% of the occasions, I was dealt a farmer's hand. A farmer's hand should not occur that often statistically.

When in a seat other than dealer, I have no trump 33.8% of the time. This should not happen.

I have lost 71% of the games. Statistically, even a very poor player should win somewhere between 45% and 55% of the time over a 100 game streak.

This pushed my rating down about 30 points.

This sort of thing makes me wonder just how random their algorithm is.

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

Unread post by Richardb02 » Sat Aug 22, 2020 9:57 am

What I have learned is that random generators are not, by definition, random. Most of them are a derivative of the Fibonacci sequence! So they are derived from a wave sequence, That is ok, the sequence parallels our universe. (The Fibonacci sequence is very interesting and worthy of an internet search.)

The clumping of cards when shuffled by humans is reflected to a certain degree by the random generator. My experience matches RedDukes. The waves are longer and without randomness with online games. Karman seems no different to me, because I have seen similar patterns on other online systems. OE posts have suggested that it takes 400 hands to have a representative sample of Euchre hands. That meshes with the results from 3,000 hands that I have compiled. I suggest that 100 hands is not enough of a data base to reflect that deals are random in any Euchre situation.

Wes (aka the legend)
Posts: 1541
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Sat Aug 22, 2020 2:05 pm

Streblerm wrote:
Thu Aug 20, 2020 11:22 am
It is Karman. I do like the play and I'll probably continue. I'd been playing A-Star for a long time. No problem finding partners, clean gameplay, hoyle rules/stick the dealer.

I've tried several euchre apps and nothing compares to karman for me. Great look, great feel and so user friendly. Reminds me of playing on PartyPoker back in the day. Now that said, I can't stand the go under rule. Adds sooooo much more variance to the game which necessarily detracts from an expert's edge. The weird loner rule doesn't really bother me that much. If anything it makes the game more fun to me becuz you can now go alone from the 3rd Seat-1st round way more often. And 2 Seat loners are easier too.
Streblerm wrote:
Thu Aug 20, 2020 11:22 am
It might be paranoia but I never felt like the deals were really random, or maybe they are too random.


Becuz we are all working with glitchy biological software that's literally designed to see patterns where there are none, I wouldn't take these thoughts seriously. When my brain starts thinking this stuff I just ignore it. And bottom line, it doesn't really matter anyways. Your job is still the same: MAKE PLAYS!! AND MAKE THE RIGHT PLAYS!! :-)

Maybe one day they'll have a euchre app as clean and fun as the Karman games app but with real euchre rules. One can dream. Also if you like the Go Under variation introduce it to others when you play in real life. Most people love it even tho I hate it!

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

Unread post by RedDuke » Sun Aug 23, 2020 12:45 am

Richardb02 wrote:
Sat Aug 22, 2020 9:57 am
What I have learned is that random generators are not, by definition, random. Most of them are a derivative of the Fibonacci sequence! So they are derived from a wave sequence, That is ok, the sequence parallels our universe. (The Fibonacci sequence is very interesting and worthy of an internet search.)

The clumping of cards when shuffled by humans is reflected to a certain degree by the random generator. My experience matches RedDukes. The waves are longer and without randomness with online games. Karman seems no different to me, because I have seen similar patterns on other online systems. OE posts have suggested that it takes 400 hands to have a representative sample of Euchre hands. That meshes with the results from 3,000 hands that I have compiled. I suggest that 100 hands is not enough of a data base to reflect that deals are random in any Euchre situation.
A Fibonacci sequence is a sequence in which the next number is the sum of the two numbers before it. I'm not totally sure what you're referring to here.

Computer software cannot generate random numbers on its own. It uses some sort of external input (the time and date, the exact second that you hit a key, whatever) as a seed - this is called entropy. It needs to use some sort of unpredictable data. The other way to do it is to use some seed and an algorithm to generate a number that appears to be random but really isn't. I suspect that what the Karman app is doing is using the second option because there's not a whole lot of need to use pure entropy to do the trick. A computer program cannot literally just roll a dice and get a totally random figure. This explains things like the pattern that I pointed out where I can lose far more often than what would happen if I was playing with actual humans shuffling the cards and dealing them out.

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

Unread post by Richardb02 » Sun Aug 23, 2020 5:04 pm

The source of my information, is software engineers, with knowledge of how the random generators operate internally. Maria, the owner/ programmer of WoCG, was a recent source that stated that same information.

I also have anecdotal experience from computer training, long ago. The random generator, at that time created a random list from 1 to 100. That taught me that the seed, had to be unique. I agree 100% with your information about the seed.

Apparently, with the off the shelf random generator software available, there is not an affordable program that can reflect the 45,000 different hands available in Euchre. Our on line games simply reflect that reality. So we accept the limitations. It is what it is.

Wes summed it up concisely: “Your job is still the same, MAKE PLAYS!! AND MAKE THE RIGHT PLAYS!!! His comment merely reflects the older wisdom, you have to play with the cards that are dealt to you. The advice even applies to life in general. My responsibility is to make the best, and constantly improving, decisions.

Isn’t there a quote from Hoyle, “thou shall not complain about the cards, that the euchre gods have dealt thee?”


Streblerm
Posts: 24
Joined: Fri Mar 22, 2019 10:58 am

Unread post by Streblerm » Mon Aug 24, 2020 10:12 am

Wes (aka the legend) wrote:
Sat Aug 22, 2020 2:05 pm

Streblerm wrote:
Thu Aug 20, 2020 11:22 am
It might be paranoia but I never felt like the deals were really random, or maybe they are too random.


Becuz we are all working with glitchy biological software that's literally designed to see patterns where there are none, I wouldn't take these thoughts seriously. When my brain starts thinking this stuff I just ignore it. And bottom line, it doesn't really matter anyways. Your job is still the same: MAKE PLAYS!! AND MAKE THE RIGHT PLAYS!! :-)

Maybe one day they'll have a euchre app as clean and fun as the Karman games app but with real euchre rules. One can dream. Also if you like the Go Under variation introduce it to others when you play in real life. Most people love it even tho I hate it!
I 100% agree that you don't change your strategy based on "streaks." The suspicion about A-star is based on my experience after thousands of games on A-star. Loaners and farmer hands are way too common, close games rarely happen. Many of the strategies discussed on this site just don't seem to work. It's a shame because gameplay is clean and it's straight rules/stick the dealer.

Granted, I've only played 100-ish games on Karman. It "feels" more like playing with humans and actual cards, which is something I've also done thousands of times.

I dislike both going-under and the loaner rule. They disturb the balance of the game IMO.

Post Reply