Taking notes during competitions?

[PeterGustafsson]

Hi!


A few weeks ago, I coached two kids from my club in a regional event. To get more hard data, I took notes of what happened during their bouts. Apart from general just looking at what happened and commenting to them what I thought that they did well and what should be fixed, I also filled in two diagrams:
1. A stem-leaf diagram which showed during which 30-second interval the hits occurred
2. A scatterplot which showed where the scoring point was, in relation to the piste lines, when it hit the opponent. Different symbols for my kids and their opponents.

Both of these produced interesting trends, especially the 2nd one - more on that later in this thread.

However, the main point of this thread is: what kind of notes do you take, as a coach? If they are more formalized, can you describe them and what information you tend to get from them?


Have a nice time!

Peter Gustafsson

[Wafath]

Originally Posted by PeterGustafsson

If they are more formalized, can you describe them and what information you tend to get from them?

I seem to recall that UNC's Coach Ron Miller's PhD thesis was on a a fencing action shorthand for taking notes. You may want to investigate it.

W

[Fenceguy]

I have used this scatter plot method for years with my son and his saber fencing. We have had pretty good results from the data. We use this information to find where his overall weakness and strengths on the strip are and train for these findings. We can also then plan future strategy against the same opponent. Many American Football coaches use a similar tracking method to find out where such things as 'pass completions' are successful and not (ie completions to the flat, long, short, ect.).

Good luck and if anything it is fun to track with your fencer.

[eac]

I think the solution is to completely abandon all paper, for all time, and take HD high-speed video. Then, slam it into some custom-written video processing software, and have a lackey (or yourself) tag each event (read: play) with the actions involved, who scored, the score and time remaining at the time, who the participants were, what yard-line it was, and stuff like that, just like they have for football. (Obviously some of those things can be automatically done by software.)
Then you can have a big searchable database, assuming you have lots of hard drives (which are a small fraction of the cost of the camera), and really be much more together in life.

This is undoubtedly the best way to analyze anything to do with fencing (or maybe any analyzable sport), although it might be overkill for lower levels. (You don't need video to see that your fencer is losing because his footwork is all over the place and he can't parry.)

It might only work for weapons with frequent lights on the box (MF/MS/WS), because there are a lot of significant things that happen between the lights in e.g. WF.

Incidentally, coaches can do this for football because they have video that actually shows the relevant parts of the field (the safeties). Reporters and fans don't have this video, so they have to resort to doing crude aggregate statistics which don't really reveal causality. I worry that the previously mentioned fencing scatterplot is more like these latter statistics, which is why I prefer to just go all-out with video, rather than looking at possibly intermediate results.

[bob46]

Originally Posted by eac

...assuming you have lots of hard drives...

oorrrr a 2 terabyte hard drive. also, does fnet's dictonary not include the word "terabyte?" its giving me a red squiggly line.

[Morale Officer]

Originally Posted by eac

This is undoubtedly the best way to analyze anything to do with fencing (or maybe any analyzable sport), although it might be overkill for lower levels. (You don't need video to see that your fencer is losing because his footwork is all over the place and he can't parry.)

I don't think this is necessarily true. It really depends on how people learn. For some people (even at the lower levels), they may actually need to see themselves in action and see what they're doing wrong (even if it's that "they can't parry") in addition to drills and verbal instructions. I am actually one that needs to see what I'm doing in addition to everything else (although it's extremely tough for me to find someone willing to come along to tape me). It's just how my brain processes information. It all depends on the person.

[Allen Evans]

psssttt.....dartfish.

[eac]

Originally Posted by Morale Officer

I don't think this is necessarily true. It really depends on how people learn.

I believe the original question was how should a coach analyze fencer problems, which is a very separate step from how the coach should proceed to attempt to get the fencer to fix said problems. It's likely that things a coach learned from video (of other fencers, even) would show up only in convoluted ways in lessons.

As for dartfish, it's so expensive they won't even tell you how expensive it is, which is a very bad sign.

[Morale Officer]

Originally Posted by eac

I believe the original question was how should a coach analyze fencer problems, which is a very separate step from how the coach should proceed to attempt to get the fencer to fix said problems. It's likely that things a coach learned from video (of other fencers, even) would show up only in convoluted ways in lessons.

As for dartfish, it's so expensive they won't even tell you how expensive it is, which is a very bad sign.

It still comes down to how people process information and learn, apply, and teach. What I quoted was you saying that video taping wouldn't help someone at a low level. I disagree because sometimes you need to see what's wrong. Not only that, but you can also catch things that you may have missed by taping - not only that, but you can rewind and see exactly what was going on (even frame by frame if you really want to) before and after a particular notable event. Besides, if you look at the big picture between the coach and athlete, it may very well be helpful to combine both the "note taking" aspect as well as the video taping. Just my thoughts on it.

[KD5MDK]

Originally Posted by eac

I think the solution is to completely abandon all paper, for all time, and take HD high-speed video. Then, slam it into some custom-written video processing software,

You mean Dartfish?

[edew]

Originally Posted by eac

I think the solution is to completely abandon all paper, for all time, and take HD high-speed video. Then, slam it into some custom-written video processing software, and have a lackey (or yourself) tag each event (read: play) with the actions involved, who scored, the score and time remaining at the time, who the participants were, what yard-line it was, and stuff like that, just like they have for football. (Obviously some of those things can be automatically done by software.)
Then you can have a big searchable database, assuming you have lots of hard drives (which are a small fraction of the cost of the camera), and really be much more together in life.

This is undoubtedly the best way to analyze anything to do with fencing (or maybe any analyzable sport), although it might be overkill for lower levels. (You don't need video to see that your fencer is losing because his footwork is all over the place and he can't parry.)

It might only work for weapons with frequent lights on the box (MF/MS/WS), because there are a lot of significant things that happen between the lights in e.g. WF.

Incidentally, coaches can do this for football because they have video that actually shows the relevant parts of the field (the safeties). Reporters and fans don't have this video, so they have to resort to doing crude aggregate statistics which don't really reveal causality. I worry that the previously mentioned fencing scatterplot is more like these latter statistics, which is why I prefer to just go all-out with video, rather than looking at possibly intermediate results.

There is a need for the writing in that the fencer (or coach) might want to record his or her thoughts from the bout. That's not something you can get from a video replay. However, it might be possible to do both or have the fencer/coach get interviewed by the camera person.

[Gav]

Does anyone know if there's good iPhone/iTouch note taking software (aside from the built in Notes app) that's good for this sort of thing?

[aamct2]

Originally Posted by Gav

Does anyone know if there's good iPhone/iTouch note taking software (aside from the built in Notes app) that's good for this sort of thing?

I should think it fairly easy to create one along the lines that PeterGustafsson described in his original post.

-- Time During Bout When Hit Occurs --
Have Start and End buttons. When you press the Start the bout timer starts ticking.

Then have buttons that correspond to touch for and touch against (or touch left and touch right). When you press those buttons, it records the touch and the time it occurred. The timer stops ticking after you press a Touch button. When you hit the start button, it picks up where the timer left off.

Then when you hit the End button, it save it as a new Note in the Notepad program, sends it to an e-mail, some other means of outputting the results.

-- Adding Where Hit Occurs Functionality --
Add a strip of buttons that correspond to the strip. After you press the Touch button to record the time of the touch, you press the appropriate distance along the strip and it adds location information to the touch.

-- Additional Features --
Add built in graphing of the results, saving of the results, comparing against other bouts, etc. Perhaps build a Desktop companion application that allows for further analyzing of the bout, overlaying it with video footage, etc.


I don't have a Mac, so I can't develop this app myself. But anyone is welcome to use this design.

~aamct2

[griffindm]

Originally Posted by Allen Evans

psssttt.....dartfish.

or if you do not have unlimited money

MotionView Software

About $125 for a complete system annual subscription.

[darius]

The USFA did something very similar to your idea for Mobile Windows 6. I'm on the beta test team, but my experience with the application was not 100% positive enough to use in real-time. The actual hit tracking was OK, it was the database entries / lookups which took too much time.

Once you have the data, it's pretty nifty, though.

We've done notation exercises at camps before, and done properly, they can yield a TON of information about the bout. I'm going to agree with eac that video should be superior, but only if shot well -- I've seen many that only have one fencer in the frame, or have bad angles, etc.

darius

[Jason]

Originally Posted by darius

The USFA did something very similar to your idea for Mobile Windows 6.

Mobile Windows crashing market share makes it borderline irrelevant. Why is the USFA wasting resources developing something for it?

[oso97]

Originally Posted by Jason

Mobile Windows crashing market share makes it borderline irrelevant. Why is the USFA wasting resources developing something for it?

The development efforts were started prior to the crash in market share. Significantly before.

[PeterGustafsson]

Hi!

Originally Posted by darius

We've done notation exercises at camps before, and done properly, they can yield a TON of information about the bout. I'm going to agree with eac that video should be superior, but only if shot well -- I've seen many that only have one fencer in the frame, or have bad angles, etc.

darius

This is what I was looking for. I agree that video has problems - I would guess that prominently among them is the fact that, in many competition layouts, it is impossible for the videographer to be in a position where good camera angles can be had. Refs, other pistes, fencers, building structure can interfere with field of vision also. Not to mention the cost of several cameras if the team has multiple fencers on piste at the same time.

I was more looking for some note taking system which a parent, who has never fenced him/herself, can be taught to manage within 30 minutes tops. The jotted-down notes should provide raw data that is useful for a coach that has not seen the bout that was noted, since the coach was observing another fencer.

At that competition, I was running from one piste to the other, observing the two fencers in our club, under-13 MxE, with gender-segregated DE bouts.

One of those fencers had a very distinct pattern. In her poule bouts, some 80-ish hits were scored. Of those, only 8 happened in a position where the scoring tip was behind the middle line. Of those 8, she lost 2-6. Of the hits which occurred where the scoring tip was near, or behind, the opponents warning line she scored a majority. Of those that were scored close to the opponent´s guard line, though, she lost a majority.

Time-wise, she had a negative scoring differential during the seconds 0-30, a positive one during seconds 31-60, and slightly negative beyond that.

All in all, she had a slightly negative scoring differential, and ended up 3T out of a field of 5.

What is your take on that, if any? I will withold mine for the moment, so that I do not prematurely influence you.


Have a nice time!

Peter Gustafsson

[Foil.Leicester]

I've got my own method which I've started using which I think might fall into what you're thinking Peter:

I jot out a rough diagram of the piste, mark one fencer with X and one with O, when one scores a touch, mark without appropriate symbol where on the piste it happened. If I'm feeling advanced and really paying attention I'll also add what the action was in my own shorthand.

A - Attack
R - Riposte
C - Counter
P - Prep
F - Fleche (I consider it different enough to attack)
L - Line (including set ups from)

And then stick a tiny note of what time in the bout it happens at.

I think adding tiny details about exactly what action happened are quite useful, particularly if you look at things like the back-line, because its such a high pressure zone its interesting to see what my student's go-to move in that zone is.

[Jason]

Originally Posted by Foil.Leicester

F - Fleche (I consider it different enough to attack)

This is going to spoil a lot of your data.
Fleche is not an action. It's a movement--like lunge or extension or advance. A fleche can be used to make an attack or riposte or counter-attack or whatever.

If you note when a fleche was used, but don't record the action that was being used, the rest of your data is compromised. If you record 5 attacks, 3 ripostes, and 10 fleches, you have very little usable info because you only noted the attacks and ripostes that weren't fleches. What actions were those 10 fleches used for? Did the fencer actually make 15 attacks and 3 ripostes? 5 attacks and 13 ripostes? 10 attacks and 8 ripostes? It's all very different info.