Compilation Video: Lapkes Attack

[CyrusofChaos]

hey guys! i have the next compilation video for you. this time it is one of my favorite fencers: dmitri lapkes of belarus in a video of his very unique attack. any and all nonmusic related comments and questions are welcome. if you dont like the music and you cant figure out how to mute the video you probably arent going to learn anything from it anyway :P

the study mentioned in the video can be found below the embedded video on this page. enjoy!

[YOUTUBE]http://www.youtube.com/watch?v=WPsoKpjcYY8[/YOUTUBE]

The study that is mentioned in the video is know as Signal Detection theory. There are two options that Lapkes has when he sees a counterattack or a fake of counterattack: he can finish or hold and ignore it. If he finishes and it is a real counterattack, the study calls it a hit: Lapkes is awarded with a touch. If he thinks he sees a counterattack and finishes when it is not real it is called a false alarm and the penalty is that he goes on defense. If he does not react to a real counterattack, this is called a miss and his penalty is that his opponent gets a touch. If he ignores the fake counterattack this is called a correct rejection and his reward is that he gets to continue attacking. For Lapkes, a miss (where his opponent gets a touch) is much worse than a false alarm (where nothing happens to him immediately) and so he has pushed his threshold to one side to get more false alarms than misses. Here is a link to the study for whoever is interested.

[yearsofwisdom]

hey Cyrus, love your vids man. i was hoping you would have some more analysis of foil fencers.

[Craig]

Is this the study that you mentioned:

http://www.cns.nyu.edu/~david/handouts/sdt/sdt.html

Craig

[oiuyt]

any and all nonmusic related comments and questions are welcome.

I think this video could have used more material on how other people deal with or break Lapkes attack as well. You have a ton of footage of successful execution, but there wasn't a huge variety in what's there -- he does something very, very well and does it similarly a lot.

Dedicating some portion of the video time to highlighting weaknesses, even at the cost of some of the repetition of successful examples, would have been beneficial. I think the only failure in his fencing that you noted was in the bout against Oh, and even there you attributed it to Lapkes not making his usual adjustments.

Show what works, and why, but also show when it doesn't work and why.

And, as always, thanks for doing this, especially at such a high level of consistent quality and so regularly. Getting a weekly video fix is great.

-B

[CyrusofChaos]

Is this the study that you mentioned:

http://www.cns.nyu.edu/~david/handouts/sdt/sdt.html

Craig

yes

I think this video could have used more material on how other people deal with or break Lapkes attack as well. You have a ton of footage of successful execution, but there wasn't a huge variety in what's there -- he does something very, very well and does it similarly a lot.

Dedicating some portion of the video time to highlighting weaknesses, even at the cost of some of the repetition of successful examples, would have been beneficial. I think the only failure in his fencing that you noted was in the bout against Oh, and even there you attributed it to Lapkes not making his usual adjustments.

Show what works, and why, but also show when it doesn't work and why.

now thats the type of comment that i want more of i will do a better job of that in the future. for now let me say that usually the way to deal with attacks in general is with first level fakes early in the bout and second level fakes later. for example in the beginning lots of fake counterattacks followed by distance pulls or parry riposte, whichever you personally are more comfortable with. later, once the opponent realizes your fake is a fake, move to the second level: fake a counterattack, fake getting out of the way (by leaning back or taking a quick step back) or fake the parry then make a real counterattack. lapkes is no different, in fact that sequence is exactly what i would recommend for dealing with people like him. the only difference i would suggest is that against him his opponent should use first level fakes for a lot longer before switching to fakes with two levels because lapkes is a very smart fencer. hope that was helpful

[oiuyt]

now thats the type of comment that i want more of

Oh, and I liked the music too...

-B

[Allen Evans]

I'll second Oiuyt here. I would have enjoyed seeing a few actions in which someone actually managed to catch Lapkes in prep or more examples of the false counter/real counter working.

Very interesting.

AE

[CyrusofChaos]

Oh, and I liked the music too...

-B

fine, ill accept that comment

I'll second Oiuyt here. I would have enjoyed seeing a few actions in which someone actually managed to catch Lapkes in prep or more examples of the false counter/real counter working.

ok i will try to include more of this in my dumitrescu video and in other videos in the future. maybe i will also make a two minute "fail" edition for the lapkes video. we will see

[springdon]

This is great, very interesting. Something I have seen in many of your commentaries (and those of others you host at fencingvideos.net) is the importance of having a game rather than just going out there and winging it. It seems clear, the more I see your videos, that different fencers are focusing on different sets of actions to eliminate variables -- trying to force your opponent to take one of the big three defensive actions (parry, retreat, counterattack) off the table and leaving a smaller set of possibilities.

For example, it seems that Lapkes has decided to hold, basically to take parry off the table (since he is often hitting so late that he can go around or though the parry when they have to choose a line.) He then also eliminates retreat as a good plan, since he is not going to finish short on an opponent who is simply retreating (since he seems very comfortable waiting until he is right on the opponent to finish.)

As you said, he has simplified his attacking game (in most cases) to forcing his opponents into a low-percentage parry game or going to counterattack (which as you said, he is expecting and has a game-theoried plan how to respond to.)

I know that this is only paraphrasing what you laid out, but it is interesting to think about how going with one action (or set of actions) is designed to limit an opponents options. It seems what a lot of the best fencers are doing, and the game is to figure out how they are trying to box you and how to get out of the box.

That is one reason that a "fail" video would be great. It seems like --going from what you laid out -- that one plan would be throw the fake counter, get him to fall short and then punish him on defense. Basically, since he has such a solid attack, don't give him ROW. Is that a common response?

Has Lapkes done the same kind of work eliminating options on his defensive game?

[Superscribe]

I have some feedback.

I like videos describing actions and describing tactics. What you are doing is taking some characteristics of a fencer and elaborating on them. I find more value in epeelion's analysis of the entire bout, and how a particular fencer may stick with, change, or abandon their usual game. For example following how Lapke normally does his holding attack, and how it develops throughout a bout with Oh, the Korean fencer. For instance, you mentioned that Lapke did very poorly against Oh, but you maintained that Lapke's style of holding till the last minute was a good choice against people who threw out a lot of blade actions on the defense? (correct me if i got that wrong) That seems wrong. Did he change? Did he adapt? How does Lapke create situations where he is able to encourage his opponent to counter-attack so that he can better use his strength (holding attack)?

At 2:30, i think there are some actions where Lapke gets counter attacked, and a few where he recieves an AiP? I wasn't clear, and during those actions if you could throw in the result (who got the touch) it'd be very helpful.

Also, you mentioned that because of Lapke's strategy, he faces a high risk of missing, but a low risk of counter-attacks. I think you mean AiP. he actually get's counter-attacked all the time, but it's okay, because he finishes.

I appreciate all the effort you've been putting into this.

[CyrusofChaos]

to springdon:

in my opinion the game is the most important thing. the game is what creates anomalies such as one fencer beating another easily while losing to someone who could beat the other one. every person knows what they like to do more often than others and what things he or she is willing to sacrifice in order to (hopefully) minimize the touches scored against them. these thresholds change from time to time for some fencers, from bout to bout for others, and still others change their games at certain points during a bout. like all fencers, lapkes also does something like this on defense. like buikevich, he has a timing counterattack, meaning he does not so much set up as wait for a time when he thinks the opponent cannot finish. he uses his athleticism to get away from the other attacks when he doesnt have a chance to counterattack, and tries to take the blade when he can. sometimes he sets up a first level fake parry riposte which works very well because of the rest of his game but wouldnt necessarily work for me because my defense is so different

lapkes's attack isnt so good that most people refuse to give him right of way because, as i said, he does not like getting counterattacked and so a first level fake will work well on him in the beginning of the bout and later second level fakes will work well too. its all just different parts of the game.

to superscribe:

yes, i also find the bout commentaries useful and i intend to do some of those when i have the time as well. these videos help me a lot too, in fact that is where i got the idea from: i spliced together a bunch of lopez counterattacks and was writing down my observations about them and figured that other people might get something out of the video that i made. so i will be doing a few more of those (at least) before doing bout commentaries, but be sure they are coming

i only showed successful examples of lapkes's attack in the video so every touch went to him. except for the one in the opening where he actually got parried but it doesnt look like it the way i edited the video

yes i mean attack in preps. the main form of criticism that i have gotten from people is that i use my own terminology and not a standard set of terms. those are the terms that make sense to me: they are what i have always used and will always use. if you need clarification, by all means ask what i mean by something. i try to be very consistent so once you learn my "language" it should be pretty easy to see what i mean

as for the effort, i love fencing and i love doing this so its no big deal although a craftsman is always please to hear his work is appreciated so thank you

[Coldfire]

The FIE youtube channel has a bout where Lapkes loses to Reshetnikov
http://www.youtube.com/watch?v=M82vY5sINh8
Looks like Lapkes stop going in with a plan after the break and Reshetnikov punishes his rushing in the middle. I'd be interested in hearing what other think.

[Mihail]

The FIE youtube channel has a bout where Lapkes loses to Reshetnikov
http://www.youtube.com/watch?v=M82vY5sINh8
Looks like Lapkes stop going in with a plan after the break and Reshetnikov punishes his rushing in the middle. I'd be interested in hearing what other think.

Off topic, but in that video, in the bout in the background, starting at about 5:15, you can see an example of the kind of crazy "points in line" that are called nowadays.

[erooMynohtnA]

http://www.youtube.com/watch?v=OgE0t...eature=related At 1:10.

At least his arm was straight.

[Mihail]

Good call. I got as good a shot of it as I could:

[CyrusofChaos]

The FIE youtube channel has a bout where Lapkes loses to Reshetnikov
http://www.youtube.com/watch?v=M82vY5sINh8
Looks like Lapkes stop going in with a plan after the break and Reshetnikov punishes his rushing in the middle. I'd be interested in hearing what other think.

well it looked to me like lapkes was attacking the same way he usually does and reshetnikov was dealing with it basically the same way i said higher up on this thread: he was faking a counterattack and then either pulling lapkes short or stepping in with a parry. he happened to get all of the parries that he stepped into on defense. he also did the counterattack with the close-out once. i thought the main problem in this bout was more due to lapkes not having enough say in the middle than anything else

[catwood1]

Off topic, but in that video, in the bout in the background, starting at about 5:15, you can see an example of the kind of crazy "points in line" that are called nowadays.

It seemed like that line call was right. Couldn't quite see, but it seemed right from a distance to me...

Idk what the deal was with the bout in the background, but it didn't seem like it was going too well for the referee. He was going to the video on almost every call in the 2nd half of the bout, and had a number of calls overturned on replay.

[Mihail]

There was no blade contact on that action, as you can see in the video Anthony linked to. And you can kinda "see," via extrapolation, from my screen cap that Pastore disengages the blade way off target, before Limbach even makes an attack, let alone a search.

[CyrusofChaos]

so apart from the comments that you have all given already is there anything else you want to see me do differently in the next video? it seems like the main thing is show more instances of it failing. is that correct?

[springdon]

Well, not to speak for everyone, but I think the best would be if you could find some middle ground between the "looking at one move" videos and the "showing the progression of one bout" videos. I think both are great, the problem is that in the first instance the videos seem to be without context and in the second case sometimes it is hard to find a bout that shows a lot of chess playing (for example, the Covaliu v Smart match there is not much happening, and the Mokretsov v Lopez is a study in Mokretsov not changing.) Also, in the full bout examples, sometimes there are really just a couple of touches that are really interesting when someone makes a change.

So . . . assuming that makes any sense to anyone but me -- a suggestion:

Rather than going with a failure video, maybe examine on move or style but add some analysis about how someone might try to adjust to negate this (whether it works or not.) For example, you could pull out those "change moments" from different bouts when someone tries to switch things up. So, given Lapkes' attack, what have different opponents done to try and address his attack? Whether they work or not, seeing people try to make adjustments is the real payoff. This would certainly also make it possible to include some things like the Oh bout as a kind of "the dangers of Lapkes game is if x."

I think one of the best parts is when you put yourself in the place of a fencer. So we might get something like, "there are a couple of ways to deal with Dumi. Here Tarantino decides to y . . ." or "Dumi sometimes gets tired or confused when his opponents do x" or "Lapkes game depends on a director who will see it when he does y"

Anyway, I know we all appreciate the effort and your time, so thanks for asking for feedback.