Forums

OK very grumpy - issues with Opening Explorer

wollyhood

I was looking at the options under this c.c opening explorer, and I get to about move 10 and according to this database there are only 2 games recorded and both say black wins 100%. But when i move that exact (black winning 100%) move in the analysis, it says in two different games, white has two different moves databased and the chance of winning is 70% to white in both games, there are still only 2.

What the actual!

This makes me REALLY grumpy and I would like someone to explain why this opening explorer is working correctly, when this evidence would suggest otherwise.

I really wanted to make a move in this particular game today but am very interested in what people say here, first.

EDIT ok yeh I did remember wrong here. There weren't 2 games after the said move there were suddenly 6. But there were only two moves for white. Anyway it doesn't change the problem that will make more sense once you see the screenshots.

destroyer8470

Alright so let me explain why when it says that there is only 1 game and there are more, its because there has only been one position just like that, but when you move the piece, there have been several positions in that position!

You'll see this when you transpose, if you transpose from one opening to another, you'll see the amount of games will change dramatically, the explorer shows games from that position.

An example you can check out is, 1.Nf3 Nf6 2.Nd4 Nd5 <--In this positions there are no recorded games. but if you go 3.e4 e5 4.Nf3 Nf6 you'll see there are lots of games.

The explorer shows games that are from that 'position' not games that follow the exact same moves. So when changing, it can change in its win ratio, in that position it was dead lost, but in this 'new' one it is not wink.png

Hope this helped happy.png

petitbonom

Id consider Opening Explorer here as a ' guide ' only.

One problem is that many games transpose , especially and obviously in the first few moves . There are many examples where a game move with only one continuation played suddenly introduces many games ( hence the transposition) .

Never rely on Explorer , always check other databases. ( Daily games only of course)

wollyhood
destroyer8470 wrote:

Alright so let me explain why when it says that there is only 1 game and there are more, its because there has only been one position just like that, but when you move the piece, there have been several positions in that position!

You'll see this when you transpose, if you transpose from one opening to another, you'll see the amount of games will change dramatically, the explorer shows games from that position.

An example you can check out is, 1.Nf3 Nf6 2.Nd4 Nd5 <--In this positions there are no recorded games. but if you go 3.e4 e5 4.Nf3 Nf6 you'll see there are lots of games.

The explorer shows games that are from that 'position' not games that follow the exact same moves. So when changing, it can change in its win ratio, in that position it was dead lost, but in this 'new' one it is not

Hope this helped

Thanks Des, but no there are the same number of games it is just who wins them that has completely changed!

wollyhood
petitbonom wrote:

Id consider Opening Explorer here as a ' guide ' only.

One problem is that many games transpose , especially and obviously in the first few moves . There are many examples where a game move with only one continuation played suddenly introduces many games ( hence the transposition) .

Never rely on Explorer , always check other databases. ( Daily games only of course)

You haven't really answered my question/ issue here, but thanks for your input. I know there are other opening exps out there that do not change their mind for no apparent reason like this one does.

Honestly, it looks like input error. Am not sure how that would be possible though, it should all be an application doing all the counting and percentages.

destroyer8470
wollyhood wrote:
destroyer8470 wrote:

Alright so let me explain why when it says that there is only 1 game and there are more, its because there has only been one position just like that, but when you move the piece, there have been several positions in that position!

You'll see this when you transpose, if you transpose from one opening to another, you'll see the amount of games will change dramatically, the explorer shows games from that position.

An example you can check out is, 1.Nf3 Nf6 2.Nd4 Nd5 <--In this positions there are no recorded games. but if you go 3.e4 e5 4.Nf3 Nf6 you'll see there are lots of games.

The explorer shows games that are from that 'position' not games that follow the exact same moves. So when changing, it can change in its win ratio, in that position it was dead lost, but in this 'new' one it is not

Hope this helped

Thanks Des, but no there are the same number of games it is just who wins them that has completely changed!

Ah well it could be coincidence that there are the same number of players, but check who played the games, they could have changed, if you could send me the line of moves you've put in I could investigate and tell you what is happening wink.png

Martin_Stahl

Yeah, showing the line may help answer what is going on.

Ilampozhil25
wollyhood wrote:
destroyer8470 wrote:

Alright so let me explain why when it says that there is only 1 game and there are more, its because there has only been one position just like that, but when you move the piece, there have been several positions in that position!

You'll see this when you transpose, if you transpose from one opening to another, you'll see the amount of games will change dramatically, the explorer shows games from that position.

An example you can check out is, 1.Nf3 Nf6 2.Nd4 Nd5 <--In this positions there are no recorded games. but if you go 3.e4 e5 4.Nf3 Nf6 you'll see there are lots of games.

The explorer shows games that are from that 'position' not games that follow the exact same moves. So when changing, it can change in its win ratio, in that position it was dead lost, but in this 'new' one it is not

Hope this helped

Thanks Des, but no there are the same number of games it is just who wins them that has completely changed!

well then you are getting confused. a few games were added/deleted(because of transpositions), the winner changed, so it seems weird. look at the games list and the games would have totally changed. so its not the result of a game changing, its just a whole another game taking its place, which had a different result.

Ilampozhil25

i think

Snookslayer

wollyhood

grrr at new zealand worship jokes this thread is about chess. I will show you the exact results of O.E.

 

wollyhood

OK so I am hesitantly going through c.c O.E. trying to play the best moves for white, while making the best moves for me black.

i have to be careful as I'm in a game here! Please don't help me, only look at this just from this move d5 and then Qc2. Of course I make dumb mistakes, I'm a noob.

so the ONLY difference between that screenshot, and this next one, is that I'm going to CLICK Qc2. THAT'S IT.

now all of a sudden white in all probability is winning by far!!! SUPER ANNOYING. These 4 games where white has won have just sort of dropped out of the ether as if they didn't exist until now. Looks like input error, doesn't it?

 

wollyhood

PLEZ EXPLAIN. This isn't the only time this has happened, it's just one of the most strikingly irregular times.

wollyhood

Can I ask yours? I would have more joy. Although, after the output of some of what they have produced - perhaps not. Perhaps standing at tomb stones asking non chess players WOULD be better.

wollyhood
Ilampozhil25 wrote:
wollyhood wrote:
destroyer8470 wrote:

Alright so let me explain why when it says that there is only 1 game and there are more, its because there has only been one position just like that, but when you move the piece, there have been several positions in that position!

You'll see this when you transpose, if you transpose from one opening to another, you'll see the amount of games will change dramatically, the explorer shows games from that position.

An example you can check out is, 1.Nf3 Nf6 2.Nd4 Nd5 <--In this positions there are no recorded games. but if you go 3.e4 e5 4.Nf3 Nf6 you'll see there are lots of games.

The explorer shows games that are from that 'position' not games that follow the exact same moves. So when changing, it can change in its win ratio, in that position it was dead lost, but in this 'new' one it is not

Hope this helped

Thanks Des, but no there are the same number of games it is just who wins them that has completely changed!

well then you are getting confused. a few games were added/deleted(because of transpositions), the winner changed, so it seems weird. look at the games list and the games would have totally changed. so its not the result of a game changing, its just a whole another game taking its place, which had a different result.

You seem to have inadvertently pointed out the major problem here.

The previous moves had to exist to get the latest moves! The games were in the database at latter moves when not in the database in the previous move! That's crap, isn't it? That's rubbish. There is something really wrong with the explorer for this to be able to happen. PLUS, at least one of these results is clearly WRONG. So it's pumping out wrong info. That's not cool.

wollyhood

I think I should delete the screenshots after I get some clued up responses. Otherwise some other noob is going to bust in here and tell me how bad my moves up to the last moves are and try and fix the game. Will check back in about 10 hours when I have my chess brain on.

Hawkeye373

It's about transpositions. The next position wasn't reached by moving Qc2

For example if you look at openings like 1. e4 Nc6 there are less than 10k games

But  2. Nf3 you see that there are more than 200k games with e5 next. And that is because most games have 1. e4 e5 2. Nf3 Nc6 

So basically it shows you positions on the board regardless of the path to get there. 

Ilampozhil25
wollyhood wrote:
Ilampozhil25 wrote:
wollyhood wrote:
destroyer8470 wrote:

Alright so let me explain why when it says that there is only 1 game and there are more, its because there has only been one position just like that, but when you move the piece, there have been several positions in that position!

You'll see this when you transpose, if you transpose from one opening to another, you'll see the amount of games will change dramatically, the explorer shows games from that position.

An example you can check out is, 1.Nf3 Nf6 2.Nd4 Nd5 <--In this positions there are no recorded games. but if you go 3.e4 e5 4.Nf3 Nf6 you'll see there are lots of games.

The explorer shows games that are from that 'position' not games that follow the exact same moves. So when changing, it can change in its win ratio, in that position it was dead lost, but in this 'new' one it is not

Hope this helped

Thanks Des, but no there are the same number of games it is just who wins them that has completely changed!

well then you are getting confused. a few games were added/deleted(because of transpositions), the winner changed, so it seems weird. look at the games list and the games would have totally changed. so its not the result of a game changing, its just a whole another game taking its place, which had a different result.

You seem to have inadvertently pointed out the major problem here.

The previous moves had to exist to get the latest moves! The games were in the database at latter moves when not in the database in the previous move! That's crap, isn't it? That's rubbish. There is something really wrong with the explorer for this to be able to happen. PLUS, at least one of these results is clearly WRONG. So it's pumping out wrong info. That's not cool.

Wrong. They were there, just in a different move order so you couldn't see them.

Ilampozhil25
Hawkeye373 wrote:

It's about transpositions. The next position wasn't reached by moving Qc2

For example if you look at openings like 1. e4 Nc6 there are less than 10k games

But  2. Nf3 you see that there are more than 200k games with e5 next. And that is because most games have 1. e4 e5 2. Nf3 Nc6 

So basically it shows you positions on the board regardless of the path to get there. 

This example. the games which magically appeared were there, just in a different move order. They were there in the previous moves. Just in 1.e4 e5 2.nf3 and not 1.e4 nc6 2.nf3

wollyhood
Ilampozhil25 wrote:
Hawkeye373 wrote:

It's about transpositions. The next position wasn't reached by moving Qc2

For example if you look at openings like 1. e4 Nc6 there are less than 10k games

But  2. Nf3 you see that there are more than 200k games with e5 next. And that is because most games have 1. e4 e5 2. Nf3 Nc6 

So basically it shows you positions on the board regardless of the path to get there. 

This example. the games which magically appeared were there, just in a different move order. They were there in the previous moves. Just in 1.e4 e5 2.nf3 and not 1.e4 nc6 2.nf3

ohhhhh got you - penny has dropped. Thanks for both the other comments adding weight to the penny.