People would ignore it if it did. On Chess24, half the kibitzers were 1200 players who would watch the eval like a hawk, totally ignore the position, and proclaim that "So and so blundered!" whenever the eval changed by at least 0.2. Apparently trash talking GM's is the patzer's wet dream. That and pretending that they thought of the computer moves themselves.

It led to funny results sometimes when the eval was wildly wrong. I remember one Bishops of opposite colors ending where Black was like 3 pawns up, but it was clearly a dead drawn ending. But the eval said -4.4, so the patzer brigade kept going on and on about how winning Black was. People who knew better told them that he wasn't, and explained exactly why he wasn't (the pawns were blockaded). They just ignored it and kept going on about how Black was winning. Then when the game was finally drawn they were scratching their heads wondering how he blew it. They hadn't looked at either the game or anyone else's comments, they were just in their own little bubble.

I am sure here is not the first to wonder at the power the evo bar has over people. However it so does have one good point.

Go through the final positions on chess.com, for example the Norway open. https://www.chess.com/events/2025-n...

and you are looking for the evo bar round about the middle but a 'White Won' or 'Black Won' scored.

This either tells you it was a loss on time or someone has resigned in a drawn position. As in the round 4 game which is now on here K Guretchii vs T Lewandrowski, 2025

The evo bar is showing an equal game but also 'Black Won.'
Kirill Guretchii had 3 minutes left on his clock https://www.chess.com/events/2025-n... so it does look like a resignation in a drawn position.
(or maybe his mobile rang, someone noticed he was wearing jeans, someone was thumping boards in the main event and he went off to see what all the fuss was all about.)

I sometimes see that and wonder how do they know and realise they must have an engine ticking over in the background. Fair enough.

Though quite a few here call it as they see it on the board in a live game and you do get a lot of interesting suggestions from others that have wee holes or holes a bus could drive through.
Most of mine are two move traps I'd expect either player to fall into. These are two bus holes.

I have to be honest I do this more because it's a fanbase fight than related to actual chess.

I concur with <Petrosianic> that eval bars probably don't do the actual chess quality any good, but it does bridge the gap for lower rated players to sort of be able to follow the flow of a game, whereas before it was just mostly one commentator vomiting 20-move lines that are more often than not plain wrong.

Of course, there's a difference between an 800, 2000 or 2400+ looking at said eval bar. The 800 will just claim 'x or y is or not better', the 2000 might calculate some concrete lines and understand the position overview, whilst the 2400+ and GMs generally understand what's going on... saying that, and often even Howell (as 2700) engines himself to the best move in dynamically complex positions also.

They're fully entitled to enjoy the game in such a manner.

Even if it means they don't really know what's going at times.

The problem is that it also misinforms them. Or at least they draw the wrong conclusions from it.

This game is a perfect example. If you just look at the eval throughout the game, you'd probably think that Black was winning this game easily, but blew it. When you tried to figure out why, you'd probably think that Black just overlooked 53. Rxe2+.

The truth is very different. In fact, even when Black was way ahead, this game was a Minotaur's labyrinth of difficult lines and traps. People think that the bigger an advantage someone has the easier it is to find the right moves. No, that's often not so. As for 53. Rxe2+, Magnus surely saw that, but believed that the Knight wouldn't be able to stop his pawns.

And that's the problem. In a public forum people don't use it as a learning tool, they use it to show off and pretend they spotted mistakes themselves, even if they aren't really mistakes.

It was especially bad on Chess24, because that site provided a very <weak> engine for non-paying customers. To give you an example, suppose the eval is +2.00. It sounds like White is winning, right? The engine would even say the words "White is winning". Okay, but don't just look at that. Instead, actually <play> out the moves that the engine is recommending.

I expect one day AI will add a note as to why they think White or Black has a plus. I do not mean the pidgin English waffle that comes with commercial programs, but a valid reasoning.

The point being made here is that people who DO use the engines often don't know what's going on, and spread misinformation to others to boot.>

The point here is that no-one has disputed anyone's right to use <fishie>, Rybka or whatever software of their choice in the privacy of their home; but <pedant> must, perforce, have the last word, so deliberately took this tack.

<Naka--Gukesh> after 23.Kg2

click for larger view

Quick, what's the eval?

I would have guessed about +0.5, maybe ~0.9-1.0 at the most from a quick scan. But those are both underestimates, and I noted so after consulting an engine:

<<sfod> It's an instructive position, I was surprised at how much an advantage White has.
...
<sfod> ("I was surprised at how much an advantage White has." - maybe shows how much more instruction I need!)>

Nakamura vs D Gukesh, 2025 (kibitz #6)

Of course it takes a master to play the position correctly and convert the advantage. Which Naka did, playing perhaps his best game of the tournament.

Just went through the game again.

From the above diagram I'd call it even. White can kick off the opposite Bishops attack with h4 - that was played but looking at it and Bb6 and Qc5 who is attacking who.

I'm pretty sure I would not have allowed Black a check on f2 or g1. Also I'd have banged in h5 a lot sooner and hoped it works.

There a bit of bluff juggling going on as Nakamura worked out his way in. Still in the pot as far I'm concerned till White got in on the back rank with a Rook Check.

Even then you would be thinking Black must have perpetual chances.

50.Bd5! Yes a very good game.

As a companion piece to the puzzle from <stone free or die>, consider the position in the game Nakamura vs D Gukesh, 2025 after Black's 24th move:

click for larger view

and then consider the position if that game had instead proceeded 20...hxg6 21.Rxb8 Rxb8 22.g3 Rf8 23.Kg2 Bb6 24.h4 Qc5:

click for larger view

If someone uses an engine and it highlights the blunder a poster should at least try to explain why. Instead you sometimes see a 'how could they play like this comment.' etc..