Topic Rybka Support & Discussion / Rybka Discussion / Advantage: black + ExtraPawn versus white + OpenLines
![[de]](/mwf/v12/flags/de.png "Germany")
I think, it is an advantage to have a pawn more than my opponent, though he has the white pieces and he has some open lines right from the beginning of the match.
But how big is this advantage?
If we want to compare with the well Know 'I play white', what can we say:
These Advantages are nearly equal?
or: the Advantage of have an extra-pawn ist greater?
or something else?
Was the advantage for Jaan really in each games bigger than 'playing WHITE in just normal matches'?
Quap
But how big is this advantage?
If we want to compare with the well Know 'I play white', what can we say:
These Advantages are nearly equal?
or: the Advantage of have an extra-pawn ist greater?
or something else?
Was the advantage for Jaan really in each games bigger than 'playing WHITE in just normal matches'?
Quap
Our computer vs. computer tests showed that the two edge pawn handicaps (a2 and h2) were only slightly bigger advantages for Black than playing White in normal chess. But the other six pawn handicaps were much larger handicaps, about four to five times as big as just playing White. This makes sense, because a pawn is considered equal to three tempi in the books, and White's normal advantage is just 1/2 a tempo. So playing Black with a pawn plus should be 5/6 of a pawn whereas White should be 1/6 of a pawn. Since the missing pawn does save a development tempo in most cases (but hurts king safety in f2 and g2 cases), perhaps a fair average value for the non-edge pawns is a bit less than 5/6 of a pawn, so the ratio of between four and five agrees perfectly with this theoretical calculation.
![[us]](/mwf/v12/flags/us.png "United States")
How do you get the calculation (behind the other six pawns) of "four to five times as big as just playing white"?
And what does the sentence "So playing Black with a pawn plus should be 5/6 of a pawn whereas White should be 1/6 of a pawn." , mean?
It seems to be an incomplete statement and confuses me somewhat.
Do you mean "Black with a pawn plus should have a 1/6 pawn advantage and "White with a pawn plus should have a 5/6 pawn advantage?
And what does the sentence "So playing Black with a pawn plus should be 5/6 of a pawn whereas White should be 1/6 of a pawn." , mean?
It seems to be an incomplete statement and confuses me somewhat.
Do you mean "Black with a pawn plus should have a 1/6 pawn advantage and "White with a pawn plus should have a 5/6 pawn advantage?
![[ee]](/mwf/v12/flags/ee.png "Estonia")
It is pure and simple mathematics. If a pawn (normally valued as 1) is considered equal to 3 tempos, and white's normal advantage is just 1/2 a tempo, then value of starting White equals 1/6 of a pawn (1/6=(1/3)/2). White, having 1 pawn less (-1) has still value of 1/2 a tempo, that equals 1/6 of a pawn. Therefore advantage for Black in this situation is 1-1/6=5/6.
If Black has a pawn less, then advantage for White is 1+1/6=7/6.
If Black has a pawn less, then advantage for White is 1+1/6=7/6.
Thanks now I understand. I understood the mathematics (I knew where the 5/6 and 1/6 came from) just not the english of what he was saying.
But I seriously doubt that a non edged-pawn in general on average is worth 3 tempi (I don't care who wrote the book).
If a tempo is a full move ahead of time (2ply ahead of time) then obviously, in most cases, 3 tempi are worth far more than a mere pawn (even a center one).
For example,
give a GM three extra tempo in the opening against another GM of the same playing strength and watch him crush the other way more times then just giving him an extra pawn in the opening.
There are many sound gambits for white that gives up a pawn at the advantage of gaining less than 3 tempo (maybe 1-2tempo).
Two and a half extra tempi is just overkill at GM level (but not 5/6 of a pawn).
Another thing is that it is very possible to develop at least one move faster with a pawn missing (as shown by Rybka against Elverest) than without. This is another 1/2 tempo that should by added. Plus add the increased mobility and opens lines and files for more freedom of movement for attack and then it is easily seen that black's advantage with pawn plus is much less than 5/6 of a pawn (and not a little). A better estimate for a non edge pawn is about 2 tempi and a better estimate of black's advantage for white missing a non edge pawn is 1/4-1/3 pawn advantage (or 1/2 - 2/3 tempo advantage), assuming at GM level.
But I seriously doubt that a non edged-pawn in general on average is worth 3 tempi (I don't care who wrote the book).
If a tempo is a full move ahead of time (2ply ahead of time) then obviously, in most cases, 3 tempi are worth far more than a mere pawn (even a center one).
For example,
give a GM three extra tempo in the opening against another GM of the same playing strength and watch him crush the other way more times then just giving him an extra pawn in the opening.
There are many sound gambits for white that gives up a pawn at the advantage of gaining less than 3 tempo (maybe 1-2tempo).
Two and a half extra tempi is just overkill at GM level (but not 5/6 of a pawn).
Another thing is that it is very possible to develop at least one move faster with a pawn missing (as shown by Rybka against Elverest) than without. This is another 1/2 tempo that should by added. Plus add the increased mobility and opens lines and files for more freedom of movement for attack and then it is easily seen that black's advantage with pawn plus is much less than 5/6 of a pawn (and not a little). A better estimate for a non edge pawn is about 2 tempi and a better estimate of black's advantage for white missing a non edge pawn is 1/4-1/3 pawn advantage (or 1/2 - 2/3 tempo advantage), assuming at GM level.
Actually I have run many computer vs. computer tests to check this theory of pawn=three tempi, and in general I find it rather accurate. If anything, in these games the pawn is worth a tad more.
Maybe in human vs. human play the correct figure is about 2 1/2. If we consider gambits for tempo, such as Smith-Morra, Blackmar-Deimer, and Goring, White gains two tempi (counting the removed center pawn as a tempo, since Black needs to move that pawn to develop and White doesn't) and so has a 2.5 tempo lead for the pawn. However statistics show that all of these gambits are at best equal for White, maybe a tad worse, so I think this confirms the 2 1/2 figure.
When computer plays human, we should take the average, so maybe 2 3/4 tempi = pawn. So pawn handicap as White is 9/11 = about 0.8 pawn, minus the average value of the opened lines. If we call that half a tempo (for the non-edge pawns), that reduces the handicap to about 0.6 pawn, which I think is the fairest value I can come up with for the six non-edge White pawns as handicap. This is between three and four times the value of the first move, a bit less than the results of computer vs. computer testing indicate.
Maybe in human vs. human play the correct figure is about 2 1/2. If we consider gambits for tempo, such as Smith-Morra, Blackmar-Deimer, and Goring, White gains two tempi (counting the removed center pawn as a tempo, since Black needs to move that pawn to develop and White doesn't) and so has a 2.5 tempo lead for the pawn. However statistics show that all of these gambits are at best equal for White, maybe a tad worse, so I think this confirms the 2 1/2 figure.
When computer plays human, we should take the average, so maybe 2 3/4 tempi = pawn. So pawn handicap as White is 9/11 = about 0.8 pawn, minus the average value of the opened lines. If we call that half a tempo (for the non-edge pawns), that reduces the handicap to about 0.6 pawn, which I think is the fairest value I can come up with for the six non-edge White pawns as handicap. This is between three and four times the value of the first move, a bit less than the results of computer vs. computer testing indicate.
![[hu]](/mwf/v12/flags/hu.png "Hungary")
BTW - for those of you who haven't put everything together yet, this is how the randomizer was born. :)
Vas
Vas
Topic Rybka Support & Discussion / Rybka Discussion / Advantage: black + ExtraPawn versus white + OpenLines
Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill