Is online poker flawed, fundamentally? - Page 29 - Gambling and Probability

Two Plus Two Forums Other Topics Probability

Is online poker flawed, fundamentally?

Post Reply Subscribe

...

Page 29 of 50

First

9 19 24 25 26 27 28 29 30 31 32 33 34 39 49

Last

Page 29 of 50

First

9 19 24 25 26 27 28 29 30 31 32 33 34 39 49

Last

03-09-2018 , 02:18 PM

#701

EvilGreebo

banned

Join Date: Nov 2013 Posts: 3,526

I have updated the google drive folder.

The code base has been updated to show the latest code. The key sequence of interest there is:

Code:

            for (int i = 0; i < 10000; i++)
            {
                deckCount += 1;
                decks.Enqueue(shuffler.Shuffle());
            }

            // 10000 times draw a randomly picked deck anywhere from position 1-500 from the front and log its top card
            for (int i = 0; i < 10000; i++)
            {
                int skipCount = new Random().Next(500);
                for (int j = 0; j < skipCount; j++)
                {
                    Deck throwAway = decks.Dequeue();
                    decks.Enqueue(shuffler.Shuffle()); // add new deck to queue as we pull old deck off
                }
                Deck ourDeck = decks.Dequeue(); // Here's the deck we want
                decks.Enqueue(shuffler.Shuffle()); // add new deck to queue as we pull old deck off
                Card card = ourDeck.NextCard();
                counter[(int)card.Suit, (int)card.Value] = counter[(int)card.Suit, (int)card.Value] + 1;
                cl1.AppendLine(card.ToString());

            }

(Note: The above had counting and comment code removed but the main functionality is shown - full code is in the code folder.

As you can see - create 10k decks, then pick a # from 1-500 and throw away that many decks, then take the next one and record the output, adding the same total number of decks to the queue.

I went thru a few iterations of file naming but settled on the one with year month day hour minute second followed by output type - so you can see each run easily sorted by name. Older iterations are in the drive though so you can review them as well.

And I think you'll find the distribution of cards is pretty much normal. In my last iteration, with random skipping, the least occurring card showed up 153 times and the most 223.

Quote

$25m Guaranteed WPM on CoinPoker

Join the action now

Daily Rewards • Splash Pots • CoinRaces

03-09-2018 , 03:46 PM

#702

pkdk

banned

Join Date: Apr 2015 Posts: 689

Quote:

Originally Posted by EvilGreebo

I have updated the google drive folder.

The code base has been updated to show the latest code. The key sequence of interest there is:

Code:

            for (int i = 0; i < 10000; i++)
            {
                deckCount += 1;
                decks.Enqueue(shuffler.Shuffle());
            }

            // 10000 times draw a randomly picked deck anywhere from position 1-500 from the front and log its top card
            for (int i = 0; i < 10000; i++)
            {
                int skipCount = new Random().Next(500);
                for (int j = 0; j < skipCount; j++)
                {
                    Deck throwAway = decks.Dequeue();
                    decks.Enqueue(shuffler.Shuffle()); // add new deck to queue as we pull old deck off
                }
                Deck ourDeck = decks.Dequeue(); // Here's the deck we want
                decks.Enqueue(shuffler.Shuffle()); // add new deck to queue as we pull old deck off
                Card card = ourDeck.NextCard();
                counter[(int)card.Suit, (int)card.Value] = counter[(int)card.Suit, (int)card.Value] + 1;
                cl1.AppendLine(card.ToString());

            }

Thank you for doing the tests I will now spend a couple of days looking at the data to look for anomalies.

Results of 10,000 deck shuffle, picking a deck at random (from 1 to 500 in the line), picking 1st card from that deck, and refilling the queue.
Total decks created: 1189498

Club-Ace : 204
Club-Two : 190
Club-Three : 172
Club-Four : 193
Club-Five : 197
Club-Six : 172
Club-Seven : 185
Club-Eight : 192
Club-Nine : 187
Club-Ten : 182
Club-Jack : 198
Club-Queen : 187
Club-King : 191
Diamond-Ace : 207
Diamond-Two : 174
Diamond-Three : 210
Diamond-Four : 207
Diamond-Five : 199
Diamond-Six : 168
Diamond-Seven : 159
Diamond-Eight : 207
Diamond-Nine : 201
Diamond-Ten : 183
Diamond-Jack : 205
Diamond-Queen : 180
Diamond-King : 194
Heart-Ace : 183
Heart-Two : 197
Heart-Three : 179
Heart-Four : 173
Heart-Five : 208
Heart-Six : 183
Heart-Seven : 191
Heart-Eight : 233
Heart-Nine : 202
Heart-Ten : 161
Heart-Jack : 199
Heart-Queen : 202
Heart-King : 197
Spade-Ace : 195
Spade-Two : 169
Spade-Three : 201
Spade-Four : 187
Spade-Five : 209
Spade-Six : 204
Spade-Seven : 198
Spade-Eight : 197
Spade-Nine : 185
Spade-Ten : 207
Spade-Jack : 189
Spade-Queen : 198
Spade-King : 209

Quote

03-09-2018 , 04:05 PM

#703

EvilGreebo

banned

Join Date: Nov 2013 Posts: 3,526

I have a question for the statisticians and probability experts following this thread...

Given n number of drawn cards, what would be the right formula for determining whether the count of a given card (for instance, 209 K

out of 10,000 draws) is within an acceptable deviation?

I was thinking of adding this to my code.

Quote

03-09-2018 , 04:12 PM

#704

EvilGreebo

banned

Join Date: Nov 2013 Posts: 3,526

Also - pkdk - I'm adding some extra functionality to allow me to set the number of runs desired - so if I wanna do a million draws instead of 10k, I can do so easily.

Quote

03-09-2018 , 04:32 PM

#705

EvilGreebo

banned

Join Date: Nov 2013 Posts: 3,526

Aaaaand for grins I'm running a million draws both ways.

Only took 3 minutes to do the sequential draw. So far the random jumping draw has been going for 14 minutes and is 31% complete.

Quote

03-09-2018 , 04:55 PM

#706

NewOldGuy

Pooh-Bah

Join Date: Mar 2009 Posts: 5,934

Quote:

Originally Posted by EvilGreebo

out of 10,000 draws) is within an acceptable deviation?

I was thinking of adding this to my code.

Calculate the standard deviation first, and then divide that into how many your count is offset from the expectation to get a Z score. That can then be converted to the probability of such a result.

Then you have to to a chi square or other test on the collection of results. I can do it later but no time now.

Edit: for 10,000 cards we expect 2500 spades. The standard deviation is sqrt(NPQ) = 43. So about 68% of the time we should be between 2500 +/-43. And about 95% of the time we should be between +/-(43*2). And about 98% of the time we should be between +/-(43*3).

For the top card to be an ace on 10K random decks we expect 10K/52 = 192. The standard deviation is 14. So about 98% of the time we expect to have between 150 and 234 aces on top. And the other 2% we expect to be over 234 or under 150.

Take the count for every pile and check the distribution using the chi square or a t test.

Last edited by NewOldGuy; 03-09-2018 at 05:18 PM.

Quote

03-09-2018 , 05:24 PM

#707

whosnext

Carpal \'Tunnel

Join Date: Mar 2009 Posts: 6,788

To follow-up a bit on NOG's post:

The number of occurrences of a given card (like the King of Spades), under the null hypothesis of random generation, obeys the Binomial Distribution.

Let P = prob of occurrence in a single trial ("success"); for a single card P = 1/52, of course
Let S = total number of successes
Let N = total number in sample

Of course, the mean of S is given by:

E(S) = N*P

The variance of S is given by:

Var(S) = N*P*(1-P).

Standard Deviation is the square root of the Variance.

SDev(S) = Sqrt[N*P*(1-P)]

In large enough samples, and 10,000 certainly qualifies, the binomial distribution can be accurately approximated by the Normal Distribution with same Mean and Variance.

We can use the known characteristics of the Normal Distribution to say how many of our observed S's should be (under the null hypothesis of pure randomness) within 1, 2, 3, 4, etc., standard deviations of the Mean.

P.S. The above should not be construed to imply that I think that testing pure counts is the best or only way to investigate OP's claims. But it would be a fairly easy first step.

P.P.S. As NOG has pointed out above, our sample represents a collection of counts that, by construction, are not independent. So doing 52 different "counts tests" is not exactly kosher. The entire sample should be tested as a whole, which is what a "Chi squared test" does. (But, as a crude first pass, the above should be okay.)

Last edited by whosnext; 03-09-2018 at 05:34 PM.

Quote

03-09-2018 , 07:38 PM

#708

pkdk

banned

Join Date: Apr 2015 Posts: 689

Results of 10,000 deck shuffle, picking 1st card from each deck.
Club-Ace : 179
Club-Two : 200
Club-Three : 171
Club-Four : 190
Club-Five : 182
Club-Six : 193
Club-Seven : 178
Club-Eight : 194
Club-Nine : 195
Club-Ten : 198
Club-Jack : 174
Club-Queen : 194
Club-King : 202
Diamond-Ace : 204
Diamond-Two : 198
Diamond-Three : 183
Diamond-Four : 172
Diamond-Five : 177
Diamond-Six : 192
Diamond-Seven : 196
Diamond-Eight : 185
Diamond-Nine : 172
Diamond-Ten : 182
Diamond-Jack : 199
Diamond-Queen : 181
Diamond-King : 218
Heart-Ace : 205
Heart-Two : 217
Heart-Three : 188
Heart-Four : 206
Heart-Five : 190
Heart-Six : 202
Heart-Seven : 208
Heart-Eight : 193
Heart-Nine : 194
Heart-Ten : 194
Heart-Jack : 180
Heart-Queen : 156
Heart-King : 191
Spade-Ace : 200
Spade-Two : 217
Spade-Three : 180
Spade-Four : 206
Spade-Five : 192
Spade-Six : 184
Spade-Seven : 207
Spade-Eight : 201
Spade-Nine : 191
Spade-Ten : 191
Spade-Jack : 208
Spade-Queen : 190
Spade-King : 200

Results of 10,000 deck shuffle, picking a deck at random (from 1 to 500 in the line), picking 1st card from that deck, and refilling the queue.
Total decks created: 1189498
Club-Ace : 204
Club-Two : 190
Club-Three : 172
Club-Four : 193
Club-Five : 197
Club-Six : 172
Club-Seven : 185
Club-Eight : 192
Club-Nine : 187
Club-Ten : 182
Club-Jack : 198
Club-Queen : 187
Club-King : 191
Diamond-Ace : 207
Diamond-Two : 174
Diamond-Three : 210
Diamond-Four : 207
Diamond-Five : 199
Diamond-Six : 168
Diamond-Seven : 159
Diamond-Eight : 207
Diamond-Nine : 201
Diamond-Ten : 183
Diamond-Jack : 205
Diamond-Queen : 180
Diamond-King : 194
Heart-Ace : 183
Heart-Two : 197
Heart-Three : 179
Heart-Four : 173
Heart-Five : 208
Heart-Six : 183
Heart-Seven : 191
Heart-Eight : 233
Heart-Nine : 202
Heart-Ten : 161
Heart-Jack : 199
Heart-Queen : 202
Heart-King : 197
Spade-Ace : 195
Spade-Two : 169
Spade-Three : 201
Spade-Four : 187
Spade-Five : 209
Spade-Six : 204
Spade-Seven : 198
Spade-Eight : 197
Spade-Nine : 185
Spade-Ten : 207
Spade-Jack : 189
Spade-Queen : 198
Spade-King : 209

Ok at first glance it seems a bit odd, I have marked every element that is under 170 down, in the random pick of deck test, there is 4 under 170 where in the ordered sets there is only 1.

Could we please repeat the same test looking for just the number results of each test to see if this is a repeat process?

Last edited by pkdk; 03-09-2018 at 07:50 PM.

Quote

03-09-2018 , 08:03 PM

#709

Lego05

Carpal \'Tunnel

Join Date: Mar 2006 Posts: 25,682

Why 170?

Quote

03-09-2018 , 08:20 PM

#710

VBAces

veteran

Join Date: Dec 2007 Posts: 2,521

I ran chi-squared tests on both samples, and neither of them is statistically different from what you would expect by chance. It makes no sense to simply find the highest or the lowest values and see if they are outside some expected range, unless you picked those particular cards beforehand because you thought they'd be different. I doubt you predicted that the 7 of diamonds would be particularly low. Since there was no reason to predict that, the 7d is just one of 52, and should not be treated specially.

Quote

03-09-2018 , 08:49 PM

#711

pkdk

banned

Join Date: Apr 2015 Posts: 689

Quote:

Originally Posted by Lego05

Why 170?

Because at a glance at the first results I stated about 170-200 average.

The second results, the results that were lower stood out to me.

Quote

03-09-2018 , 08:52 PM

#712

pkdk

banned

Join Date: Apr 2015 Posts: 689

Quote:

Originally Posted by VBAces

I have not looked at the sequences yet either. It just an anomaly I spotted in the comparison test of the two sets of results.

If 4 are under that means 4 or more are over.

I need a second comparison to see.

1 and 4 are obviously a difference, if we ran the test several times and they all showed something similar, that shows an anomaly.

Quote

03-09-2018 , 09:07 PM

#713

whosnext

Carpal \'Tunnel

Join Date: Mar 2009 Posts: 6,788

Surely you realize that you will be unable to give your hypothesis a legitimate test without using formal statistical tests. And these formal statistical tests will require complex formulas that will need to be programmed (or the data fed into a program designed for the purpose).

There is no way that anybody could fairly draw any conclusion, one way or the other, based upon inspecting the raw counts by eye. And nobody could, would, or should be swayed by any conclusions drawn by inspecting the raw counts by eye.

The Chi-squared test, probably the best test for the situation, is not complicated. But to someone unfamiliar with many probability and statistical concepts, it may appear to be essentially a "black box".

If you have the time, I suggest looking up the formula for the Chi-squared test of frequencies (also called the Pearson Chi-squared test). I would import the counts into a computer program/framework and calculate the Chi-squared test statistic.

Repeat this on a large number of 10,000 datasets. Or on a small number of 1,000,000 datasets.

It should become clear if the claim of "repeat value bias" is confirmed or refuted.

Quote

03-09-2018 , 09:57 PM

#714

VBAces

veteran

Join Date: Dec 2007 Posts: 2,521

I have a feeling you can do this as many times as you'd like and the tests are going to show that these are exactly what you would expect by chance. I'll be happy to run the chi-squared every time someone comes up with results, but I'm pretty sure they are going to show that there is no bias in which cards come up, regardless of how the test is accomplished.

Quote

03-09-2018 , 10:47 PM

#715

EvilGreebo

banned

Join Date: Nov 2013 Posts: 3,526

Results of 1,000,000 deck draws (counts only)
Sequential Draw:

Code:

Results of 1000000 deck shuffle, picking 1st card from each deck, each deck taken in sequence.
Club-Ace : 19025
Club-Two : 19135
Club-Three : 19363
Club-Four : 18965
Club-Five : 19265
Club-Six : 19024
Club-Seven : 19127
Club-Eight : 19360
Club-Nine : 19103
Club-Ten : 19324
Club-Jack : 19286
Club-Queen : 19048
Club-King : 19203
Diamond-Ace : 19230
Diamond-Two : 19060
Diamond-Three : 19257
Diamond-Four : 19127
Diamond-Five : 19479
Diamond-Six : 19334
Diamond-Seven : 19144
Diamond-Eight : 19187
Diamond-Nine : 19217
Diamond-Ten : 19521
Diamond-Jack : 19426
Diamond-Queen : 19173
Diamond-King : 19186
Heart-Ace : 19197
Heart-Two : 19252
Heart-Three : 19102
Heart-Four : 19270
Heart-Five : 19235
Heart-Six : 19257
Heart-Seven : 19376
Heart-Eight : 19272
Heart-Nine : 19215
Heart-Ten : 19240
Heart-Jack : 19341
Heart-Queen : 19208
Heart-King : 19190
Spade-Ace : 19452
Spade-Two : 19162
Spade-Three : 19376
Spade-Four : 19318
Spade-Five : 19394
Spade-Six : 19225
Spade-Seven : 19050
Spade-Eight : 19165
Spade-Nine : 19129
Spade-Ten : 19465
Spade-Jack : 19132
Spade-Queen : 19390
Spade-King : 19018

Random skip draw

Code:

Results of 1000000 deck shuffle, picking a deck at random (from 1 to 500 in the line), picking 1st card from that deck, and refilling the queue.
Total decks created: 92487766
Club-Ace : 19197
Club-Two : 19164
Club-Three : 19337
Club-Four : 19419
Club-Five : 19176
Club-Six : 19030
Club-Seven : 19275
Club-Eight : 19092
Club-Nine : 18903
Club-Ten : 19002
Club-Jack : 19227
Club-Queen : 19285
Club-King : 19113
Diamond-Ace : 19204
Diamond-Two : 19248
Diamond-Three : 19380
Diamond-Four : 19157
Diamond-Five : 19309
Diamond-Six : 19195
Diamond-Seven : 19452
Diamond-Eight : 19144
Diamond-Nine : 19602
Diamond-Ten : 19384
Diamond-Jack : 19120
Diamond-Queen : 19151
Diamond-King : 19097
Heart-Ace : 19116
Heart-Two : 19270
Heart-Three : 19275
Heart-Four : 19389
Heart-Five : 19148
Heart-Six : 19061
Heart-Seven : 19306
Heart-Eight : 19352
Heart-Nine : 19285
Heart-Ten : 19460
Heart-Jack : 19234
Heart-Queen : 19289
Heart-King : 19333
Spade-Ace : 19514
Spade-Two : 19001
Spade-Three : 19249
Spade-Four : 19281
Spade-Five : 19250
Spade-Six : 19334
Spade-Seven : 19001
Spade-Eight : 19074
Spade-Nine : 19266
Spade-Ten : 19173
Spade-Jack : 19101
Spade-Queen : 19193
Spade-King : 19382

All four files are on the drive now as 20180309_151717...

Quote

03-09-2018 , 10:49 PM

#716

whosnext

Carpal \'Tunnel

Join Date: Mar 2009 Posts: 6,788

Um, 99.9999999% of the population of planet earth does not believe that there is a "repeat value bias" caused by the way that Pokerstars pre-populates its random decks into a queue.

This thread concerns the remaining 0.0000001% of the planet's population. So, to the extent that this wonderful thread is going to lead to anything productive, it may be useful to find ways for that 0.0000001% population to at least partially understand and trust the statistical evidence and tests.

It has been my lifetime experience, people are very reluctant to change a long-held view simply because someone else tells them they are wrong.

Quote

03-09-2018 , 11:05 PM

#717

EvilGreebo

banned

Join Date: Nov 2013 Posts: 3,526

pkdk if you're concerned that only 1 value was "short" in the direct deal vs. 4 in the random then you should compare 10 sets of 10k draws and see how many times that occurs for starters.

Because one outlier like this really doesn't mean anything.

Quote

03-09-2018 , 11:13 PM

#718

EvilGreebo

banned

Join Date: Nov 2013 Posts: 3,526

I'm way more rusty on this than I realized - like I remember principles but not practices, so let me translate this into dumb programmer language:

Quote:

Originally Posted by whosnext

Let P = prob of occurrence in a single trial ("success"); for a single card P = 1/52, of course
Let S = total number of successes
Let N = total number in sample

In a 10,000 draw sample where we pick 1 card each, N = 10,000 right?

Quote:

Of course, the mean of S is given by:

E(S) = N*P

E(S) = 10,000 * (1/52) = 192.3076~

The variance of S is given by:
Var(S) = N*P*(1-P).
10000 * (1/52) * (1-(1/52)) = 188.6094

Standard Deviation is the square root of the Variance.
SDev(S) = Sqrt[N*P*(1-P)] ->13.7335

Quote:

P.P.S. As NOG has pointed out above, our sample represents a collection of counts that, by construction, are not independent. So doing 52 different "counts tests" is not exactly kosher. The entire sample should be tested as a whole, which is what a "Chi squared test" does. (But, as a crude first pass, the above should be okay.)[/I]

Could you try explaining this again? I got lost here.

Quote

03-09-2018 , 11:21 PM

#719

NewOldGuy

Pooh-Bah

Join Date: Mar 2009 Posts: 5,934

Quote:

Originally Posted by EvilGreebo

If we define outlier as being more than 3 standard deviations from expectation, it would be an anomoly to NOT have about 2% of the result sets be "outliers".

Quote

03-10-2018 , 12:03 AM

#720

whosnext

Carpal \'Tunnel

Join Date: Mar 2009 Posts: 6,788

The Pearson Chi-Squared test is in virtually every basic probability and statistic textbook.

I will try to write it down here, but interested parties should do an internet search or consult a statistics book.

Suppose we have a population which can be sub-divided into K classes. In our case, K is 52 and the classes are the 52 different playing cards As, Ah, ..., 2d, 2c.

Further suppose we are interested in testing whether all of the true underlying frequencies of each of the K classes in some specific population are equal. Call this frequency P. In our case, the population in question is the cards dealt by Pokerstars and P is 1/52.

To test this hypothesis we take a large sample of size N from this population (where large means that we get a sufficiently large number of observations in each of the K classes).

Let Si be the number of sample observations in the ith class. Clearly the sum of the Si will equal N.

Clearly, the expected number of observations in each class is simply N*P or, in our case, N/52. (If N = 10,000, N*P = 192.308.) Call this Y. So in our case Y=192.308.

Then Pearson showed that the test statistic given by:

X = Sum [i=1 to K] ((Si-Y)^2)/Y

has a Chi-squared distribution with (K-1) degrees of freedom.

You then need to compare the test value you get for X above with the appropriate critical values of the Chi-squared distribution that can be found in tables in books and on the internet.

Hope this helps and hope I didn't mess up the explanation.

Last edited by whosnext; 03-10-2018 at 12:11 AM.

Quote

03-10-2018 , 07:36 AM

#721

Kelvis

banned

Join Date: Dec 2013 Posts: 8,737

You can take a horse to the water but you can't make him drink.

Quote

03-10-2018 , 07:54 AM

#722

pkdk

banned

Join Date: Apr 2015 Posts: 689

Quote:

Originally Posted by EvilGreebo

Results of 1,000,000 deck draws (counts only)
Sequential Draw:

Code:

Results of 1000000 deck shuffle, picking 1st card from each deck, each deck taken in sequence.
Club-Ace : 19025
Club-Two : 19135
Club-Three : 19363
Club-Four : 18965
Club-Five : 19265
Club-Six : 19024
Club-Seven : 19127
Club-Eight : 19360
Club-Nine : 19103
Club-Ten : 19324
Club-Jack : 19286
Club-Queen : 19048
Club-King : 19203
Diamond-Ace : 19230
Diamond-Two : 19060
Diamond-Three : 19257
Diamond-Four : 19127
Diamond-Five : 19479
Diamond-Six : 19334
Diamond-Seven : 19144
Diamond-Eight : 19187
Diamond-Nine : 19217
Diamond-Ten : 19521
Diamond-Jack : 19426
Diamond-Queen : 19173
Diamond-King : 19186
Heart-Ace : 19197
Heart-Two : 19252
Heart-Three : 19102
Heart-Four : 19270
Heart-Five : 19235
Heart-Six : 19257
Heart-Seven : 19376
Heart-Eight : 19272
Heart-Nine : 19215
Heart-Ten : 19240
Heart-Jack : 19341
Heart-Queen : 19208
Heart-King : 19190
Spade-Ace : 19452
Spade-Two : 19162
Spade-Three : 19376
Spade-Four : 19318
Spade-Five : 19394
Spade-Six : 19225
Spade-Seven : 19050
Spade-Eight : 19165
Spade-Nine : 19129
Spade-Ten : 19465
Spade-Jack : 19132
Spade-Queen : 19390
Spade-King : 19018

Random skip draw

Code:

Results of 1000000 deck shuffle, picking a deck at random (from 1 to 500 in the line), picking 1st card from that deck, and refilling the queue.
Total decks created: 92487766
Club-Ace : 19197
Club-Two : 19164
Club-Three : 19337
Club-Four : 19419
Club-Five : 19176
Club-Six : 19030
Club-Seven : 19275
Club-Eight : 19092
Club-Nine : 18903
Club-Ten : 19002
Club-Jack : 19227
Club-Queen : 19285
Club-King : 19113
Diamond-Ace : 19204
Diamond-Two : 19248
Diamond-Three : 19380
Diamond-Four : 19157
Diamond-Five : 19309
Diamond-Six : 19195
Diamond-Seven : 19452
Diamond-Eight : 19144
Diamond-Nine : 19602
Diamond-Ten : 19384
Diamond-Jack : 19120
Diamond-Queen : 19151
Diamond-King : 19097
Heart-Ace : 19116
Heart-Two : 19270
Heart-Three : 19275
Heart-Four : 19389
Heart-Five : 19148
Heart-Six : 19061
Heart-Seven : 19306
Heart-Eight : 19352
Heart-Nine : 19285
Heart-Ten : 19460
Heart-Jack : 19234
Heart-Queen : 19289
Heart-King : 19333
Spade-Ace : 19514
Spade-Two : 19001
Spade-Three : 19249
Spade-Four : 19281
Spade-Five : 19250
Spade-Six : 19334
Spade-Seven : 19001
Spade-Eight : 19074
Spade-Nine : 19266
Spade-Ten : 19173
Spade-Jack : 19101
Spade-Queen : 19193
Spade-King : 19382

All four files are on the drive now as 20180309_151717...

Quote

03-10-2018 , 08:46 AM

#723

Monteroy

Carpal \'Tunnel

Join Date: Dec 2006 Posts: 16,720

Quote:

Originally Posted by Kelvis

You can take a horse to the water but you can't make him drink.

He has no interest in drinking, and I don't think anyone believes differently. Despite all of the powerful, verifiable evidence disproving his belief - all he will do is cherry pick through the data to find some tidbits that he can cling to to further his belief.

That said, this thread actually evolved into something that usually does not happen, where some actual discussion of statistical methods and data analysis took place, along with some real data being used.

That certainly helped further expose the OP for the non-expert that he is in these areas, but that was not the underlying agenda, which made it an interesting read. In some regard - what the OP does at this point (even though we can all pretty much guess that he will find a way to keep his obsession going) does not matter.

Quote

03-10-2018 , 09:03 AM

#724

pkdk

banned

Join Date: Apr 2015 Posts: 689

Quote:

Originally Posted by Monteroy

I will accept if I am wrong when I finish analysing the data. At the moment I am looking at the numbers, so far both sets of results from both tests have shown more under's on the second test. That sir is a difference, and if further tests show this to be a repeat occurrence, then there is a problem.

Also we have not done the preliminary tests using card 1 and card 10, which again may show something.
I do not make final conclusions based on only two sets of test results.

I know exactly what I am looking for in the results. Even if I find the numbers to be ok, then I have got a long winded task of looking at the sequences.

Added- If both test are equal in results, then because of variance,

There is a 1/2 chance the first test has more under's than the second test. each time the test is run

There is a 1/2 chance the second test has more under's than the first test . each time the test is run

Upto yet 2/2 for the second test.

Last edited by pkdk; 03-10-2018 at 09:23 AM.

Quote

03-10-2018 , 09:23 AM

#725

Monteroy

Carpal \'Tunnel

Join Date: Dec 2006 Posts: 16,720

Quote:

Originally Posted by pkdk

I will accept if I am wrong when I finish analysing the data.

You will never admit you were wrong. How can you? You have dedicated a decade + to this bizarre belief. Sure, there are many people here with actual knowledge in this area doing work for you and saying very clearly (with no agenda) that your theory is not valid, and further that what you are doing with the data is an incorrect approach.

Despite this, you will find a way to make the data work for your beliefs. Nobody thinks otherwise.

Quote:

Originally Posted by pkdk

At the moment I am looking at the numbers, so far both sets of results from both tests have shown more under's on the second test. That sir is a difference, and if further tests show this to be a repeat occurrence, then there is a problem.

They have explained to you why this is not significant, but that will not stop you from believing it is significant.

Quote:

Originally Posted by pkdk

Also we have not done the preliminary tests using card 1 and card 10, which again may show something.
I do not make final conclusions based on only two sets of test results.

Your final conclusion will be to thank them for their work (you are polite in that situation) and then misinterpret their work to suit your beliefs. Even though they all knew that going in, they will eventually stop helping you, and you may interpret that as an effort to hide the truth from you, or their lack of ability to truly see your gifted vision. How often you use the word ostensible will be your choice.

Quote:

Originally Posted by pkdk

I know exactly what I am looking for in the results. Even if I find the numbers to be ok, then I have got a long winded task of looking at the sequences.

Yeah, you know what you are looking for, so you will come up with a way to find it, even when it is an invalid approach. Basically, that is the exact opposite of how to do proper research. That is a form of bias (don't recall the name) - and perhaps it is one of the ones listed in this fine song. Perhaps others can play "Bias Bingo" with your posts in the future, that would be ostensible.

https://www.youtube.com/watch?v=3RsbmjNLQkc

All the best.