Slow preparations and the attack

The last post and its conclusions (in a nutshell: start the bout in slow prep and bias towards the attack; move towards fast prep and defence in the second half) seem to have generated some interest and more than a few questions. Main groups of questions were:

  1. What datasets did you use, and plan to use?
  2. How are you classifying preparations and actions, and do you have a rigorous set of definitions for each?
  3. Really, defense?

I’m currently laid up for at least a week tending to another quadriceps tendonitis flare-up (9 months and counting), so here, have some answers and more stats.

What datasets did you use and plan to use?

We used competitions with World Cup-grade sabre fencers, even if the event itself was not a world cup. To date the dataset is small but should grow every time new competition videos become available. Bouts used to date were:

  • Asian Championships 2010 Men’s Sabre Team, Gold medal match, Korea v.s. China
  • Warsaw Grand Prix 2011 Men’s Sabre Individual, Gold medal match, Won Woo Young v.s. Alexey Yakimenko
  • Chicago World Cup 2013, Men’s Sabre Individual, Gold medal match, Szabo v.s. Szilagyi
  • London Olympics 2012, Men’s Sabre Team, Gold medal match, Romania v.s. Korea
  • Chicago World Cup 2013, Men’s Sabre Team, Gold Medal match, Russia v.s. Italy

Being as we are in Australia, we rely pretty heavily on videos from the FIE and Andrew Fischl (a.k.a CyrusofChaos) for the analysis. We do plan on wandering off to a couple of World Cups this year to grab some of our own videos as well.

How are you classifying preparations and actions, and do you have a rigorous set of definitions for each?

We’re classifying preparations and actions according the main teaching syllabus used at the Sydney Sabre Centre (aka the Codex). The terms are broadly based on those used in fencing clubs based in the USA, the UK and Australia but have been extensively revised and clarified (e.g. we make a distinction between a ‘step’ with just the front foot, and an ‘advance’ made with a front step and bringing up the back foot). When the latest version gets updated I’ll probably post the glossary to the blog, but it’s a fairly involved affair with several hundred entries.

The other part is whether the classification of actions themselves are rigorous. That’s a tougher question to answer. Most the analysis has been done by me at this stage, so at least the data are more or less consistent. The other folks who are doing the classifications are all sabre fencers who have been trained at Sydney Sabre for at least a year, so the classifications should also be broadly consistent with how I’m doing it. Upshot is that I think the data are okay, but the noise level is high and we should be careful about drawing conclusions from small statistical differences, especially before we’ve tested it on the piste.

Really, defense with fast prep?

Yes. Though our local itinerant Korean swordsman, Dong Hwan Kim, was stressing the point that we should be starting with slow prep with a bias on the attack. Ably demonstrated at a recent NSW state competition before he started messing around:

One of the notable things in this bout was how more than half the successful preparations were slow, followed by medium prep. Fast preparations made up less than 10% of the successful preparations. What’s more interesting is what the different preparations worked against: slow preparations succeeded against all other types of preparation, but fast never beat slow, and medium only worked against medium prep.

140222_prep

From slow preparation the most common winning action was AiT, and this combo made up about a third of all successful actions in the entire bout.

140222_win-prep-action

The second most common action was advance then fall short, usually from either a slow or medium prep. These occurred about a quarter of the time. Yet these combos had fairly bad odds of success – advance fall short in slow and medium prep made up about half of all unsuccessful actions as well.

140222_lose-prep-action

Less commonly attempted actions which had an excellent success rate were advance draw cut, AoP fall short and advance back step counter-attack. Why? One possibility is that each of these actions encourage the opponent to do what you want them to do, e.g. advance back step encourages the opponent to hold their attack for the counterattack, advance (extension) induces the opponent to launch an extended attack for your draw cut, and AoP/check helps your opponent finish short. By contrast, advance fall short induces your opponent to hold their attack, which isn’t great if you’re attempting to make them fall short.

So – slow preparations and attacks work. At least if you’re a former Korean national junior team fencer up against a totally outclassed opponent (which for now is pretty much all of us).

In the next few posts I’ll start laying out some early ideas about how to do decent stats analysis on actions outside of the 4m – a harder problem because of more external factors, more scope for false positive/negatives, and greater variety of parameters – and a couple of articles on the biomechanics of sabre footwork and how we’re going to get my knee back in shape before the Athens World Cup.

Fast, small, defensive: effective start tactics in sabre

Sabre fencing was once the weakest discipline of the three Olympic fencing events. It had a reputation for being a loud and boring sport where every point consisted of two guys charging at each other, then screaming at the referee to award them the point.

2011+European+Fencing+Championships+Day+Three+ThJCkhJMF4Al
Two top-grade sabreurs appealing for the point. This used to happen all the time, but is becoming rarer as the sport matures. (July 15, 2011 – Source: Dean Mouhtaropoulos/Getty Images Europe)

Since the 2008 Olympics, such simultaneous actions in sabre have become less common. There are a few reasons. Referees have become much more consistent in awarding the attack to the first person who commits to a lunge, rather than the person who runs in the quickest (or yells the loudest). Sabreurs have also become better at hitting on their opponent’s preparation. They’ve also become a lot more mobile than say 10 or 20 years ago. This all makes for some interesting exchanges at the start of the bout – people can’t just run into the middle expecting to get the attack or the simultaneous. Case in point: the men’s sabre team final at the London 2012 Olympics between South Korea and Romania. (edit: Apparently the IOC doesn’t let third part sites link to Olympics videos, even if they are on Youtube. Have some sweet World Championships action instead.)

But wait: there were still a large number of simultaneous actions at the start. Conventional theory says that the best basic tactic in sabre is still to use the simultaneous, especially at the start and end of the bout, to set up other actions. These actions should be set up with a slow preparation step (very common at the 2008 Olympics), and almost never with fast footwork in the middle (due to the risk of the opponent attacking on preparation).

So the questions on my mind are:

  1. Is the simultaneous a ‘safe’ way to set up the opponent the next action? i.e. does it have better-than-even chances of winning or drawing against a non-simultaneous action from the opponent?
  2. Are slow preparations more likely to lead to a successful action than fast or medium tempo preparations?
  3. Given (1) and (2), what are the most effective combinations of preparation and action in the 4m centre zone?

For the purposes of this exercise, I defined ‘success’ as either winning the point immediately in the 4m centre zone, or the winning the priority outside of the 4m centre zone immediately after the initial exchange of actions.

Method

I got the data by enlisting folks from the Sydney Sabre Centre to watch Youtube videos of World Cup bouts and other high-level competitions. Each point in these bouts was classified in terms of:

  • Winning preparation: the preparation (slow/medium/fast) used by the sabreur who had success in the 4m centre zone for that point.
  • Losing preparation: the corresponding preparation for the other sabreur.
  • Distance: the distance between the sabreurs when one or the other initiated their action. Distance was roughly classified as ‘deep’ = within lunge distance, or ‘short’ = advance lunge distance or greater.
  • Winning action: the action performed by the winning sabreur.
  • Losing action: the corresponding action by the losing sabreur.

Results

I did some quick and dirty frequency counts in Excel of the preliminary data (from two team bouts and six individual bouts, 244 data points). Here is the summary table for the analysis as of October 2013:

4m_actions_prob

Is the simultaneous a ‘safe’ way to set up the opponent the next action?

Generally not.

Simultaneous attacks initiated by one party (denoted as “AiT”) succeeded only 39% of the time, despite comprising around 49% of all attempted actions.

simul_win

simul_lose

Simultaneous works okay against advance fall short (i.e. a sloppy retreat without using a check) and counter-attacks, but rarely works against opponents who use a check fall short (denoted as AoP fall short), attacks on preparation (denoted AoP), and even the risky parry riposte forward.

Are slow preparations more likely to lead to a successful action than fast or medium tempo preparations?

Slow preparations are okay – but its more important to either go slow or fast than to be somewhere in the middle.

4m_preps_prob

Slow preparations were attempted less often (97 counts) than fast (173) or medium preparations (220). They were successful about half the time, whereas fast preparations were successful slightly more often (57%) and medium preparations slightly less (47%).

I’m not sure exactly why this is, but it’s interesting.

Given (1) and (2), what are the most effective combinations of preparation and action in the 4m centre zone?

win_combo

lose_combo

No easy answer here, but the most successful combos appear to be:

  • Fast prep – AoP fall short
  • Fast prep – AoP
  • Fast prep – Draw cut
  • Slow prep – AoP fall short (but surprisingly, not AoP)
  • Slow prep – Draw cut
  • Slow prep – Simultaneous: interestingly, simultaneous actions worked 50% of the time when used with slow prep, which didn’t happen with either fast or medium preps.

An important caveat is that the distance between the sabreurs is critical – some actions have very different odds of success when the distances are small (i.e. ‘deep’ distance):

win_combo_deep

lose_combo_deep

  • Fast preparation – AoP works great, though AoP fall short is less effective and draw cuts are not recommended at all.
  • Slow preparation – AoP fall short and draw cuts work, most other actions don’t.
  • Simultaneous is surprisingly effective in deep distance.

Compare situations where the distance is large (i.e. ‘short’):

win_combo_short

lose_combo_short

  • Going for the simultaneous when the opponent is far away is dangerous, regardless of whether you do a fast or slow preparation.
  • AoP is risky, given the large distances involved, but AoP fall short has good chances of success.
  • Things that feint the simultaneous, such as advance fall short and draw cuts, are also effective.

So how should I be starting?

  1. Always watch for the distance of the opponent to decide which set of actions to choose between. In practice, this probably means some degree of educated guesswork as to their closing speed and habits during the first step from the start.
  2. It’s generally safer to be far away from your opponent in the 4m than attempt to close distance = small, composed footwork.
  3. Start the bout with slow preparations, looking for either the simultaneous if the opponent is close and composed, or the draw cut/advance fall short if they are distant but closing fast. (If they stop attacking, chill out and march them down as in Sabre 101).
  4. Once you’ve settled in to the opponent’s timing, start pressuring your opponent with fast preparations. You should be pretty good at predicting their distance at this point: go for the attack-on-preparation if they get close, feint the AoP then make them fall short if they are distant, and throw in the occasional draw cut or stop hit if you can’t decide.
  5. If they start getting frantic, go back to slow preparations and grind them down.

On reflection, the above isn’t too different from what most of my coaches have been telling me for the past couple of years. One of the more intriguing anomalies is the draw cut, which was a significant part of the game played by guys like Lopez from a few years ago, but has become a little rarer in recent years. It appears to be a bit more effective than I would have expected. By contrast, advance parry riposte is making a comeback (see Chicago 2013 team finals) but doesn’t appear to have great odds for success.

Next steps are to get a few thousand more data points and test out the above in training to see if it all works – and why.

Show me the data

Science is a method to distil truth from the world. The core of science is experimentation: formulating a hypothesis about how something works, aggressively testing it, collecting observations in the form of data, analysing the results, and coming to a conclusion. All of these steps require “invention, sagacity, [and] genius” (Whewell, History of Inductive Science, 1837, and in Philosophy of Inductive Science, 1840) – and in my personal experience, a generous helping of luck.

It can be nigh on impossible to truly apply the scientific method in some fields. I have the greatest respect for people trying to work in these areas, and a fair amount of sympathy too. Sometimes it’s because no one can figure out how to actually do more than just observe, for example in astrophysics. In other cases, such as economics, ethical considerations limit the scale and types of experiments that can be performed. Then there are those fields where the results are so ambiguous as to challenge the best analytical techniques available, like psychology and nutrition.

Scientific methods are slowly becoming more common in areas far removed from the halls of academia. Evidence-based policy making in government, data analytics in business strategy, and sports analysis are three ways in which the tools and mindsets of science are being disseminated through to broader society. These are tough fields to work in, given the complexity of the systems involved, the constraints on what can be done, and time scales available; but the potential rewards are huge.

This blog documents my attempts to apply science to my day-to-day life. The idea that truth comes from experimentation is probably the one constant in my career to date – a career that has been maddeningly unfocused: I started off as an engineering undergrad in computer engineering, then branched off to medical science, cryptanalysis, photonics and surface (bio)-chemistry before going to grad school in molecular biophysics (supervised by a former bioscience head at Los Alamos). The birth of my first son knocked me off my career trajectory, whatever that might have been. These days I work as a management consultant for a cool boutique by day, and a sabre fencing instructor by night.

I’ll try to tackle three questions in this blog for the most part:

  1. What’s the most effective way to train a 29 year old, 5′ 7”, injury-prone male in sabre fencing in under 20 hours a week?
  2. How should I be raising my kids now to prepare them for the world of 2040 (including worst-case of zombie/jellyfish apocalypse)?
  3. Can a person with a laptop, free (or cheap) software and public data derive insight into how the world works?

I’ll be relying on you to keep challenging the assertions in this blog. The name, by the way, comes from a bit of wisdom from a grizzled old post-doc back when I first started in a lab:

“Trust no one. Assume nothing. Show me the data.”