WEBVTT 00:00.960 --> 00:01.800 Hi there. 00:01.980 --> 00:05.270 In this lecture, I want to talk about bias crushing of over. 00:05.280 --> 00:08.360 It's a nice play on King's crushing or King's crushing. 00:08.490 --> 00:12.060 So we want to be bias crushing in search of objectivity. 00:12.090 --> 00:17.140 That's one of the key quotations, in my view, that Alexander Alekhine made. 00:17.160 --> 00:21.540 Chess, first of all, teaches you to be objective. 00:21.690 --> 00:25.620 Now, when we say objective, that is kind of unbiased. 00:25.620 --> 00:27.810 Objectivity is like being unbiased. 00:28.020 --> 00:30.730 Not easily swayed by a particular opinion. 00:30.750 --> 00:36.840 Trying to be more evidenced based more on evidence, concrete evidence. 00:37.050 --> 00:42.580 Tigran Petrosian has mentioned that a strong player knows the rules and laws of the game. 00:42.600 --> 00:48.590 A talented player knows everything that a strong player knows and he also sees exceptions to the rules. 00:48.600 --> 00:55.140 But major test talents we call them geniuses, gradually transform these exceptions into our new rules 00:55.140 --> 00:59.100 and so on ad infinitum, since chess is inexhaustible. 00:59.190 --> 01:00.220 Inexhaustible. 01:00.240 --> 01:04.960 So Tigran Petrosian talks about the evolution of chess and our own evolution of chess. 01:04.980 --> 01:07.110 We also have this kind of process. 01:08.190 --> 01:13.890 The kinds of modernists found useful exceptions to establish theory, and they became part of the theory 01:13.890 --> 01:17.850 that you could, for example, just influence the sense of control, controlled sense. 01:17.850 --> 01:20.900 You don't need to literally occupy with pawns. 01:20.910 --> 01:26.670 So objectivity is a great thing as as a cool, you know, attacking player. 01:26.820 --> 01:29.070 We need to be objective. 01:29.250 --> 01:36.180 It is like being unbiased that objectivity, that ruthlessness the CRO mistakes do post-mortems of our 01:36.180 --> 01:36.840 own games. 01:36.840 --> 01:41.340 Finding absolute truth helps us when we play stronger and stronger opponents. 01:41.340 --> 01:48.270 We need to start with the truth, you know, as far as we can ascertain, sometimes engines don't give 01:48.270 --> 01:49.050 the truth you need to. 01:49.260 --> 01:54.120 Even the immortal game I discovered, you know, recently in constructing this course, if you don't 01:54.120 --> 01:59.340 put the engines for long enough, they think one of the moves is a mistake, as if it's, you know, 01:59.370 --> 02:00.030 a losing position. 02:00.030 --> 02:00.450 But it's not. 02:00.450 --> 02:04.710 If you put it on for more, the engine will realize the huge potential of the position. 02:04.710 --> 02:08.490 So chess is difficult even for engines to gain truth. 02:09.960 --> 02:15.090 And we see also with the neural networks like Alphazero coming in and destroying Stockfish only after 02:15.090 --> 02:16.410 a few hours of training. 02:17.100 --> 02:20.730 This is an example I'm going to show you soon in this lecture. 02:20.940 --> 02:23.670 So the truth of chess. 02:23.670 --> 02:28.290 Chess is astronomical in nature, so it's difficult to determine a truth. 02:28.290 --> 02:35.310 We only have our biases which we try and crush more and more to get a scalability more and more for 02:35.310 --> 02:36.360 stronger and stronger opponents. 02:36.360 --> 02:42.900 Ideally, if we want to improve our chess, we might just want to just love looking at attacking games 02:42.900 --> 02:43.230 as well. 02:43.230 --> 02:48.570 By the way, you don't have to be a competitive player to enjoy the games of this course, but I'm looking 02:48.570 --> 02:51.930 from the perspective that you do want to beef up your attacking skills. 02:52.470 --> 02:58.200 So when we talk about bias in research methods, when I was at Brunel, there was Brunel University 02:58.200 --> 03:04.470 in London, you know, research methods was talked about how if you take one or two samples, they might 03:04.470 --> 03:05.040 be biased. 03:05.040 --> 03:12.180 You need often greater samples and a variety of samples to reduce bias in things that you're researching. 03:12.180 --> 03:14.790 So it's a basic research ethos. 03:15.720 --> 03:22.320 And in chess, because games are recorded with PTEN, we've got more and more examples to base our conclusions 03:22.320 --> 03:30.480 on, and it's like an evolving central database on history and culture which has recorded is absolutely 03:30.480 --> 03:32.190 beautiful aspect of chess. 03:33.210 --> 03:38.310 So we get a greater and greater sample size of games to base kind of rules. 03:38.310 --> 03:41.220 The rules should be qualified more and more and more. 03:41.220 --> 03:45.480 So for example, a pawn chain, you don't have to just undermine a pawn chain and its base. 03:45.480 --> 03:48.000 You could undermine it at its head. 03:48.000 --> 03:53.820 So in the French defense, for example, if you play C6 French, you could undermine the pawn chain 03:53.820 --> 03:54.420 at its head. 03:54.420 --> 03:59.460 So there are sometimes there are no rules apart from the actual rules of chess. 03:59.460 --> 04:02.580 Basically, it really depends on the position. 04:02.760 --> 04:07.590 And attacking chess has gone through major evolutions in its history. 04:07.590 --> 04:13.650 So the romantic era of chess had a bias for beauty, for entertainment, beautiful combinations. 04:13.740 --> 04:19.020 They didn't mind about playing on something in the opening, potentially with with gambits, which might 04:19.020 --> 04:20.730 be a little bit unsound. 04:20.730 --> 04:27.870 But nowadays it's not so popular to play those romantic era gambits like The King's Gambit or the Evans 04:27.870 --> 04:33.150 Gambit, because we have computers telling us exactly how to equalize against, for example, the Evans 04:33.150 --> 04:34.500 Gambit or the King's Gambit. 04:34.500 --> 04:35.400 There are different ways. 04:35.400 --> 04:39.180 There are so many different ways of actually playing against The King's game as well. 04:39.210 --> 04:41.130 Devlin Aronian said that that's the problem. 04:41.160 --> 04:41.670 The King's Gambit. 04:41.670 --> 04:44.670 You need to know all the different counters that are possible. 04:44.910 --> 04:46.980 So there is a bit of an overhead there as well. 04:46.980 --> 04:54.120 It's not, you know, these romantic era relics were evidence that actually chess is also it's not in 04:54.120 --> 04:55.050 a vacuum. 04:55.050 --> 05:00.030 It's in the context of its historical and technological. 05:00.360 --> 05:04.320 And social and economic context. 05:04.320 --> 05:08.400 The games actually reflect the part of history that they're in. 05:08.430 --> 05:11.580 Usually they encapsulate the culture of the time. 05:12.330 --> 05:18.420 So we have a bias through the history books, not having the technology, not having the communications 05:18.420 --> 05:19.250 infrastructure. 05:19.260 --> 05:25.140 So back in the days the romantic era was was justified by itself because people didn't really know how 05:25.140 --> 05:26.760 to handle the game. 05:26.760 --> 05:28.070 It's the unsound stuff. 05:28.080 --> 05:29.030 They really didn't. 05:29.040 --> 05:32.180 So, you know, the bias a beautiful combination is playing on Sunday and opening. 05:32.190 --> 05:33.750 It was acceptable. 05:33.840 --> 05:37.680 It was an acceptable trade off, but things move on. 05:37.860 --> 05:43.200 So central occupation challenged by the so-called hyper modernists and there are lots of interesting, 05:43.200 --> 05:47.010 useful exceptions which became incorporated in chess theory. 05:47.010 --> 05:49.770 So we can play positions in very different ways. 05:49.770 --> 05:54.360 We can play hyper modern openings, we can play with finger toes, we can even, you know, double. 05:54.780 --> 06:00.000 And so there's a lot of hyper modern ideas which are absolutely wonderful. 06:00.030 --> 06:03.780 The notion of accumulation of advantages is measurable. 06:03.780 --> 06:05.130 You know, we play games, we play online. 06:05.130 --> 06:08.400 We can look at the computer often and see how the advantage has changed. 06:08.400 --> 06:10.350 The notion of central control is measurable. 06:10.380 --> 06:14.760 Even if you put a knight on the corner there, mathematically, there's a basis for it. 06:14.790 --> 06:17.970 So we're not completely saying ridiculous things. 06:17.970 --> 06:25.020 If there's a mathematical basis for things, pawns not going backwards when I started Lasker. 06:26.090 --> 06:32.180 For example, a lot of his opponents and Capablanca were playing irreversible pool moves. 06:32.180 --> 06:36.050 The notion of irreversibility is a very sound and important concept. 06:36.050 --> 06:41.840 There are very, very sound evidential concepts which are good to kind of, you know, think about that 06:41.840 --> 06:48.050 a lot of the time in Laskas time or Capablanca time or altcoin's time players were irreversibly making 06:48.050 --> 06:52.900 poor moves and not really realizing that they're the only piece you can't retract back. 06:52.910 --> 06:59.480 So there is often a mathematical or legally enforceable basis Nimzowitsch blockades, you know, is 06:59.480 --> 07:02.240 basically saying the pawn can't move if you're blockading a pawn. 07:02.240 --> 07:07.820 So the whole nimzowitsch my system, a lot of that is actually very logical and justifiable, even with 07:07.820 --> 07:08.510 great reverence. 07:08.510 --> 07:15.230 It's like certain axioms of chess, which they're not in the legal rules of chess, but they are worth 07:15.230 --> 07:22.550 thinking about from a mathematical and objectively logical perspective, where we can't get enough evidence 07:22.550 --> 07:25.850 to say it's it's bad blockade or it's good blockade. 07:25.850 --> 07:27.230 Well, you can see what it does. 07:27.230 --> 07:31.580 It's, you know, restraining the opponent's pieces, reducing counterplay. 07:31.970 --> 07:37.160 It's kind of measurable in the chessboard not to give opponents too much of a chance to come back at 07:37.160 --> 07:37.670 you. 07:37.880 --> 07:43.610 So we also had, you know, the advances of the neural networks was absolutely fascinating as a game 07:43.610 --> 07:44.210 changer. 07:44.210 --> 07:49.490 Like Alphazero coming and beating Stockfish eight at the time was absolutely amazing, and I followed 07:49.490 --> 07:51.590 avidly the Layla project. 07:51.590 --> 07:56.090 I looked at lots and lots of later games and put them on channel because I thought they they represented 07:56.090 --> 07:59.540 really interesting new ideas about how we think about chess. 07:59.540 --> 08:05.630 So things like form pawns and things about sacrificing pawns, which we thought we threw away in the 08:05.630 --> 08:06.860 romantic era. 08:07.640 --> 08:11.820 But they came back with the neural networks because they're seeing a lot more. 08:11.840 --> 08:18.110 The, the problem of the alpha beta engine approach, if that's a measure of truth, it's it's taking 08:18.110 --> 08:20.840 material as one of the main considerations for evaluating positions. 08:20.840 --> 08:25.400 And if it can't calculate how to get back the material, it often labels things as unsound. 08:25.400 --> 08:27.380 But is that the truth of the matter? 08:27.410 --> 08:28.940 Far much, far. 08:28.940 --> 08:36.290 Seeing in a way, neural networks seeing games end to end can often justify very, very deep positional 08:36.290 --> 08:44.810 pawn sacrifices, which the the alpha beta based traditional engines would underestimate quite often 08:44.810 --> 08:45.860 to their peril. 08:45.860 --> 08:52.550 So even in the engine world, the notion of truth is is kind of a function of how they actually operate 08:52.550 --> 08:57.230 the strengths and weaknesses of different engine algorithms, you know, using neural networks. 08:57.230 --> 09:02.210 So thankfully, in this course, by the way, all the game examples I've been checking with the latest 09:02.210 --> 09:05.540 like Stockfish, which has neural network capability. 09:05.540 --> 09:11.030 So we get some sort of notion of truth and objectivity to base our own attacking chess on. 09:12.200 --> 09:17.420 I like in particular when I'm playing chess to be nerdy about the downsides of the opponent's position. 09:17.420 --> 09:24.050 For me, that's I'm trading off mental effort and time to find issues objectively as I can in the opponent's 09:24.050 --> 09:24.830 position. 09:25.100 --> 09:31.100 But you need to retain the empathy that some positions are so many, there's so many traps they're worth 09:31.100 --> 09:31.910 going into. 09:31.910 --> 09:36.380 Even, you know, especially on fast the time controls, having the initiative, having the opponents 09:36.550 --> 09:42.710 in like minefield situations is having the human empathy for the practical battle that we face against 09:42.710 --> 09:44.030 other human beings. 09:44.450 --> 09:50.480 So we should retain the idea of difficulty to play or, you know, how many landmines in this position, 09:50.480 --> 09:54.050 how many is an absolute minefield of traps for the opponent? 09:54.050 --> 09:59.060 You know, then we can we can play more freely gamuts because we get a certain level of success. 10:00.170 --> 10:04.910 So okay, so we want to be biased, crushing in general in search of objectivity. 10:04.910 --> 10:11.000 So if there's one thing to learn from alekhine, just one to remember from this course, Chess, first 10:11.000 --> 10:13.100 of all, teaches you to be objective. 10:13.280 --> 10:16.400 But yeah, it's a very, very hard thing in chess. 10:16.880 --> 10:21.200 So here Alphazero against Stockfish in 2018. 10:21.200 --> 10:24.440 So a computer match around 255. 10:25.010 --> 10:26.690 So let's have a look at this position. 10:26.690 --> 10:35.360 95 was played and actually there is a tactical threat which is kind of ignored in this position. 10:35.840 --> 10:43.760 B5 was played, allows Knight takes a free check and it looks as a hold on hasn't black just won a pawn? 10:45.020 --> 10:48.590 So Alphazero just doesn't mind. 10:48.590 --> 10:53.330 It's kind of relaxed about this D4 because it's basically saying, well, that's a road to the king. 10:54.020 --> 10:58.760 Well, I'm not I'm not saying for categorically I know what alphazero thinking, but it is a road to 10:58.760 --> 11:01.640 the king is a semi-open fa which we know are pretty dangerous. 11:01.760 --> 11:04.940 So there is a kind of gambit here being played. 11:05.960 --> 11:11.120 And it is a you know, it looks pretty dangerous because we also have this dark square bishop without 11:11.120 --> 11:14.960 a counterpart, which is an attack ingredient which is emphasized in this course. 11:14.960 --> 11:19.310 So night bd7 we have Bishop E to Queen f6, the numbers should be two. 11:19.310 --> 11:24.620 So there is a convergence, it seems, of a danger between the bishop and the rook. 11:24.620 --> 11:25.250 So we have. 11:25.800 --> 11:27.850 Four, which seemed more material grabbing. 11:27.870 --> 11:33.090 So the traditional engines just looking after their material as a priority because it's easier to measure 11:33.090 --> 11:35.580 for their evaluation functions. 11:35.580 --> 11:38.820 So for saying have my --, take my pawns. 11:38.970 --> 11:43.350 So Rook Geoffrey and the is a bit more in Siberia. 11:43.350 --> 11:49.530 So we've got another attacking ingredient emphasized Queen the queen offsides semi final ingredients 11:49.830 --> 11:54.810 bishop var counterpart ingredients, team ingredients rook and bishop Maybe other team members come 11:54.810 --> 11:58.620 in rapidly while the defensive team is a bit asleep over here. 11:58.620 --> 12:02.130 So things are brewing up in terms of attacking ingredients. 12:02.340 --> 12:03.930 So castling, queenside. 12:03.930 --> 12:05.280 So the teams coming together. 12:05.280 --> 12:15.520 Bishop F3 concrete France now emerging like rook H one we have Queen H and Rook H one queen f6 and now 12:15.540 --> 12:18.450 just King B one just saying I'm I'm chilled with the position. 12:18.450 --> 12:22.950 I'm playing a quiet move here to improve prospects, getting my king a little bit safer. 12:22.980 --> 12:30.480 G6 dot square weaknesses being being created Rook Jiji one just casually accepting this position, a 12:30.480 --> 12:37.560 couple of pawns down and just trusting in the attacking ingredients, intuitively looking at the position, 12:37.560 --> 12:39.870 not just the material element of the position. 12:40.350 --> 12:47.250 So we have a full King one, so improve the king a bit more away from the opponent's light square. 12:47.250 --> 12:51.540 Bishop So this is all a dot square now, Rook G seven and now E4. 12:51.540 --> 12:59.700 So starting to open up things F4 if F takes E4 was played, Bishop takes E4, Queen f7 Bishops C one 12:59.700 --> 13:00.120 for example. 13:00.120 --> 13:03.540 This position is going to be nice for whites. 13:03.540 --> 13:08.850 So for example, if it opens up here, so it opens up here, then this is going to be trouble. 13:08.850 --> 13:09.750 After D5. 13:09.750 --> 13:15.120 You can see things are brewing up on the diagonal in some lines, but anyway, F4 trying to keep things 13:15.120 --> 13:16.020 shut down. 13:16.380 --> 13:23.940 We have C5, Queen is Rook C one of six E5 trying to break open things again for the bishop. 13:23.940 --> 13:25.020 And that's another ingredient. 13:25.020 --> 13:30.810 We have the bishop pair here as well as the bishop without a counterpart and just rookie one. 13:32.130 --> 13:34.740 So that's a very, very accurate move as well. 13:34.740 --> 13:35.790 There is accuracy here. 13:35.790 --> 13:39.750 If details are being played, it seems, you know, this would be fine for black. 13:39.780 --> 13:43.680 There'll be no problems here objectively as far as we can tell. 13:43.680 --> 13:49.260 So Rook HD one, though, pulls up the pressure and now opens up the position. 13:49.260 --> 13:55.080 Bishop takes E4 Queen feint D5 so opening up the bishop so another pawn. 13:55.080 --> 14:00.000 So right now isn't this rule breaking, you know, rule breaking. 14:00.000 --> 14:04.560 But these are not the hard rules like legal rules of chess, where rule breaking in terms of giving 14:04.560 --> 14:05.160 out all the pawns. 14:05.160 --> 14:10.590 Wasn't this isn't this a very unsound gambit system being played in this game? 14:10.590 --> 14:20.910 Free pawns down now so Bishop G4 every willing to give another pawn Bishop D7 if Bishop takes him free. 14:20.940 --> 14:22.140 You know Rook have won. 14:22.140 --> 14:27.330 We're going to be getting a pawn back with interest with that nasty pin on the Queen and we can build 14:27.330 --> 14:28.140 up our team. 14:28.140 --> 14:33.360 Pressure builds up here on F6 with interest, so we actually end up being better there. 14:33.480 --> 14:41.550 So Bishop D7, we have Queen C3 piling on the pressure amplification strategy, which is very good. 14:41.550 --> 14:46.920 Nice H five We have rook five C6 rook C one amplifying the rooks as well. 14:47.010 --> 14:48.840 So there's authority being built up. 14:48.840 --> 14:55.830 That's a great attack ingredient to build up firewall authority and amplification authority on the diagonal. 14:55.860 --> 14:57.330 We have six. 14:58.390 --> 15:06.490 We have queen the full centralising slightly C takes B five so you can see C meltdowns evidence if rookie 15:06.490 --> 15:10.570 eight her rook takes the 8/19 be six. 15:10.570 --> 15:17.080 And this position is fascinating because because of this authority know that's mate. 15:17.080 --> 15:23.050 So what happens here if if six this is almost like a zugzwang Queen's sub six what does black actually 15:23.050 --> 15:23.830 play here? 15:23.980 --> 15:31.050 Bishop C Then this rook takes E and Queen's hands G seven If H five, then believe it or not, you know 15:31.120 --> 15:36.640 there's another attacking ingredient here, not just in terms of making the King, but the past pawns 15:36.640 --> 15:37.930 lost to expand. 15:38.110 --> 15:43.480 We have Bishop A6 here it's a crusher and then B seven. 15:43.480 --> 15:45.280 And you know, black is overstretch. 15:45.280 --> 15:49.090 So a wonderful position where we start to see symptoms. 15:49.570 --> 15:51.640 You know, this doesn't look like a good move in principle. 15:51.640 --> 15:53.440 It dismantles the pawn chain. 15:53.890 --> 16:01.960 Bishop B1 is not mining all these pawns down, but who has the actual authority in the possession here? 16:02.380 --> 16:05.440 So this is material grabbing versus authority. 16:06.100 --> 16:10.360 So there's amplification, there's a foul authority, there's kind of. 16:10.450 --> 16:12.340 Yeah, there's pressure points. 16:12.340 --> 16:13.780 We have Bishop C six. 16:14.260 --> 16:20.110 If rook c eight as rook takes D5 as possible, this possession of queen takes the E5, check Queen D 16:20.110 --> 16:25.450 four and then we have Queen D six and this possession we can play Bishop A2. 16:25.450 --> 16:26.050 Ouch. 16:26.050 --> 16:28.420 Look at these absolutely wonderful bishops. 16:28.420 --> 16:35.170 So Bishop C six, Rookie six Rook have some rogue G one Look at the pressure being built up on these 16:35.170 --> 16:37.000 key squares around the King. 16:37.000 --> 16:45.460 Queen G seven Queen takes her for rookie eight rook D 697 Queen cc1 rook have six. 16:45.460 --> 16:48.160 We have now a four. 16:48.190 --> 16:53.980 And look at this slow motion attack with F5 trying to destroy and undermine the opponent's king. 16:53.980 --> 16:56.920 So here Rook takes up 6 to 9, takes another F5. 16:56.930 --> 17:00.850 Wow, what a thing to do being so many pawns down. 17:00.850 --> 17:04.660 But you know, it would take distant endgames to exploit any of these extra pawns. 17:04.870 --> 17:07.540 In the meantime, the black king is under fire. 17:07.540 --> 17:09.340 We have f tanks G6 here. 17:09.370 --> 17:14.670 This has now reached, you know, reached a desperate point, was a meltdown of King's safety. 17:14.680 --> 17:19.300 And so Queenie free and she is a desperate attempt to get rid of the queen's. 17:19.300 --> 17:23.500 But white now is is winning in this position. 17:23.500 --> 17:25.500 This is an absolutely winning position. 17:25.510 --> 17:28.040 It's a big of rook have won, for example. 17:28.090 --> 17:35.350 It's just overwhelming this position we have Knight 6h7, which is desperate. 17:35.350 --> 17:42.460 That's just giving up a piece rookie free and one is able to convert this into a win with the extra 17:42.460 --> 17:43.120 bishop. 17:43.870 --> 17:46.390 So what on earth have we seen here? 17:47.560 --> 17:53.150 You know, it's a battle of objectivity, game changer, as Natasha Ragan and Matthew said in a book, 17:53.200 --> 17:58.780 game changer how neural networks have changed kind of the rules of the game. 17:58.780 --> 18:05.560 Not not the hard, hard rules, the legal rules, but the rules of strategy, the rules of tactics, 18:05.560 --> 18:09.760 the rules of material versus authority of the possession. 18:10.360 --> 18:14.410 So that almost echoes central occupation versus central control. 18:14.410 --> 18:15.970 You want authority in the possession. 18:15.970 --> 18:19.000 You don't just want to be extra material. 18:19.360 --> 18:21.160 Material is a bit like occupation. 18:21.160 --> 18:24.190 It's about an unjust occupation, having more material, occupying things. 18:24.190 --> 18:26.560 You really want your pieces to have authority. 18:26.560 --> 18:28.840 So here, you know, black resigns. 18:28.840 --> 18:31.180 Now, this is a very, very light analysis. 18:31.180 --> 18:33.340 I just wanted to highlight to you. 18:33.370 --> 18:36.850 We're in a continual search for objectivity. 18:36.850 --> 18:43.840 Chess is such a vast, astronomically, vast game that even, you know, our relatively recent engines 18:43.840 --> 18:50.860 just a few years back are not playing to the limits of the game by any stretch, even on long time controls, 18:50.860 --> 18:53.380 There's more beautiful things to be uncovered. 18:53.620 --> 18:57.070 Chess is such a supremely rich game. 18:57.070 --> 18:59.200 So our objectivity. 19:00.620 --> 19:08.580 Needs to be kind of moderated because we play a very ridiculous fast time controls quite often. 19:08.600 --> 19:10.370 If you like online chess, like me. 19:10.370 --> 19:16.430 And so just having fun with the game, you know, having pressure points, having king attacks, that 19:16.430 --> 19:22.220 bias is acceptable because the opponents are often not able to find, you know, the objective, the 19:22.220 --> 19:27.650 sound continuations and they're using time on the clock as well, which is another piece of the chessboard, 19:27.650 --> 19:28.600 the opponent's clock. 19:28.610 --> 19:29.570 You can use it as a weapon. 19:29.570 --> 19:34.030 If you're so much up in time, that's going to be winning new games in practice. 19:34.040 --> 19:37.290 So there's a difference between practice and theory as well. 19:37.310 --> 19:44.600 But anyway, the key thing I wanted to say with this introduction, nature, it's useful to be bias 19:44.600 --> 19:48.350 crushing in general, not just in chess, but in other subjects. 19:48.350 --> 19:54.440 You know, we're continually learning all the time in science, in technology, in mathematics, in 19:54.440 --> 20:00.590 physics, especially all the arguments in physics, how new theories overtake previous theories. 20:00.590 --> 20:07.400 So this search of bias crushing in search for objectivity are tools like modern chess engines help us 20:07.400 --> 20:10.970 to become more, more kind of biased crushing. 20:10.970 --> 20:16.400 It's good to do even just quick post-mortems If you lose even online games just to sharpen things, 20:16.400 --> 20:22.280 it's often very, very useful to sharpen ideas, to sharpen opening variations, to get rid of the flaws. 20:22.280 --> 20:23.570 I know it's painful. 20:23.570 --> 20:25.510 I'll you just play this game. 20:25.520 --> 20:27.350 You want to forget about it and go to the tomb. 20:27.350 --> 20:30.620 But it's good to sometimes if you get a few minutes, just quickly review. 20:30.650 --> 20:37.040 If the longer the time it's the more painful, the more those deeper postmortems get you even more kind 20:37.040 --> 20:42.590 of ideas to reformat your game a bit, to to revisit, to reduce bias in your game. 20:42.890 --> 20:48.230 So, yes, for me, you know, chess is about being super nerdy about the opponent's position to find 20:48.410 --> 20:49.250 the downsides. 20:49.250 --> 20:55.730 I like to put in the effort the time to get those finesses of possession, to build up those small little 20:55.730 --> 21:00.560 upsides, the accumulation of little upsides, so we get a crushing attacks from that. 21:00.560 --> 21:06.440 So, yeah, bias crushing in search of objectivity is a very interesting thing that Alekhine had said. 21:06.470 --> 21:10.390 Chess, first of all, teaches you to be objective. 21:10.400 --> 21:13.310 Okay, I hope you enjoy this little lecture of things to think about. 21:14.030 --> 21:14.780 Thanks very much.