Soren Thompson on Fencing, Training for the Unknown

Soren Thompson – two-time épée Olympian and Team World Champion

by Geoffrey Loss & Jason Rogers

In this article, we talk to Soren Thompson, two-time épée Olympian and Team World Champion, about his refreshing point of view on training for fencing’s complex tactical game.

How Most of Us Think about Fencing Tactics

People often think of fencing as a complicated game of rock-paper-scissors: if I expect you to hit directly, I can parry; if you expect me to parry, you can do a feint before hitting; if I expect you to feint, I can counterattack, and so on. And this isn’t necessarily a wrong way of thinking about it. But rock-paper-scissors is a guessing game—as soon as you commit to one option, it’s very difficult to change to another. Maybe you have a sense of how the other guy likes to fence, and you’re able to guess right more often than not—but at the end of the day, your game still relies on a factor of chance.

In fencing, you have the right to do anything within the rules. What happens when you lunge at your opponent? They can parry, they can counterattack, they can lunge back at you, they can try to make you fall short and so on. In other words, there’s nothing they’re not allowed to do. Some of their choices are better than others, certainly, but there are numerous responses to your action available to them. All of these possibilities and counter-possibilities create exponential complexity. So how do you train for this?

In the lead up to the 2012 London Olympics, Soren realized this is a problem with common fencing tactics and tried to create a better solution for his fencing. He rejected the notion that the solution for the complexity of fencing is to fall back on guesswork. Instead, he trained with a specific focus on improving his ability to solve problems in real-time. By doing this, he found he could better identify and execute spontaneous solutions as challenges and opportunities presented themselves during an action.

Ok, we know that’s a lot to process, so don’t worry, we are going to unpack this in detail for you. But before we go into Soren’s approach to this important topic, we should talk briefly about some major differences in types of sports and how that impacts how you train for them.

Open Skill Sports vs. Closed Skill Sports

Soren was quick to understand that one of the main challenges in fencing is that it is what’s described as an “open skill” sport, but that it’s often taught as if it were a “closed skill” sport. We have written about this distinction in our previous article about finding focus in your practice, but  what is the difference? A gymnast performing a front handspring vault is an example of a closed skill. You have a planned routine, your environment is constant and you train by performing the same movements over and over. When you compete, you are measured on how well you perform that practiced routine.

But fencing and other sports such as hockey, judo and tennis, by contrast, primarily demand open skills. The actions that are performed must be adapted based on a changing environment and/or the actions of your opponent. You have to recognize what’s happening, sense the best action to take given the situation, and then execute it. Your success is measured by a combination of how well you do all of those things. There are no style points in fencing. If you take the world’s most beautiful parry, but your opponent finishes to a different line, even by accident, you lose that touch. In other words, unlike a closed skill sport, perfect intention and beautiful technical execution do not guarantee success. What, how, and, perhaps most importantly, when you do your actions matter and ultimately determine the outcome of your competition.

How to Win at Rock-Paper-Scissors – and Fencing

One of Soren’s insights was that, while your original intention is important, your ability (or inability) to change that decision mid-stream is actually much more so. For example, I might think that the best action is to hit straight with my lunge, but if halfway through my lunge I see that you are trying to parry, I want to be able to change, perhaps to a disengage lunge. If we’re playing rock-paper-scissors, I can plan to throw rock. But I gain a huge advantage if I’m able to switch my choice to scissors when I realize you are throwing paper.

But, Soren points out, most people don’t spend enough time training their real-time decision-making ability. In many lessons, coaches will direct your every move. For example, they might tell you to lunge and hit straight a few times, followed by a lunge and hit with a disengage, then lunge and hit with a beat, etc. This type of controlled, deliberate repetition is important when learning a technique for the first time, as is the case with beginners, but as a fencer becomes more competent, it has rapidly diminishing returns on the fencer’s ability to apply those skills successfully in a bout.

Yet often, it is still the case that in lessons for intermediate and high-level fencers, “they know how it’s going to end before it starts,” says Soren. “For me, that is the opposite of valuable. I think that it’s counterproductive because there is always the possibility for the action to be slightly different than you expect” in a real bout. Even at a high level, a lot of coaches train like this. The actions they have their students do might be more complicated and the execution might become faster and stronger, but the fencer still knows exactly what action they are going to perform (or, at most, which one of two or three actions they are going to perform).

“The easier thing to train for is the thing that you know, but often in a match you don’t know what you think you know,” says Soren. “People generally think along these lines: ‘Something happened in the past, and I’m going to risk more or less everything on it happening again in the future.’ But you don’t necessarily have an informational advantage over your opponent to make that decision. Your opponent knows what happened in the past too, and so for you to say that you are going to just pick the next logical tactical action doesn’t make sense. Your opponent is doing the same thing, and the better your opponent is, the less likely they are to do what you expect and the less likely your guess will be correct.”

Do you have some sort of intention?

Of course, Soren’s advice isn’t to go in with no forethought of any kind and just stand there and try to react. “Certainly there’s a starting point. It’s not necessarily a plan,” he says. “I would say it’s more like ‘do you have some sort of intention?’ So there's a theme in what I'm doing and a place where I might want the action to wind up, but the difference is that when I decide to do something, I don't assume that I know how the action is going to end before I start. Because when that is the case, I feel that the mind very much goes dark and you are stuck doing this initiated plan for as long as it takes to complete it, which can be a pretty long time. In the interim, the condition that you were planning for, that you were basing everything off of, may have changed. If it has changed, your great idea might have quickly turned into a terrible one. Whatever happens, I want the best chance to still be successful.”

At this point you might be thinking, “Duh, I’m obviously going to change my plan if the situation changes. That’s a no brainer.” But you must understand that because most of fencing happens so quickly that most of your decisions are completely unconscious. The sport is too fast to be able to read a situation, make a decision, and execute your plan consciously. You must be able to, for example, take a parry automatically, without having to think through what motions you need to go through like you did when you were a beginner. For this reason we must train specifically to build this skill of adaptability into our fencing, so that we are able to execute our decision making capabilities as flexibly and automatically as we can our physical and technical fencing skills.

If you’re going to the store, for example, you have to plan your route and start driving, but if halfway there you see a traffic jam ahead, you are going to need to figure out how to change routes. Even in actions that take a fraction of a second to execute, Soren trains specifically to give himself the maximum amount of time to find a better route.

Building Depth in Your Fencing Actions

You also might read this and argue that many great fencers succeed by using a more closed approach. And that’s true! There are times when it’s advantageous to have a pre-decided plan. Let’s look at the example of Daryl Homer’s last touch in the Olympic semifinal, which is one we have written about in a past article. At 14-14, he correctly anticipated his opponent, Mojtaba Abedini (Iran), attacking him to the head and committed to his action (parry five) in advance. Daryl solved and executed his action perfectly, scoring an epic touch that vaulted him into the Olympic Final.

Let’s quickly summarize again the difference between “closed skills” versus “open skills” as they pertain to fencing. Applying a closed skill approach is best described as performing a pre-planned action, whereas an open skill approach entails executing an action that evolves as the bout context changes. Soren points out that there isn’t strictly a right or wrong way to handle every scenario. Rather, there exists a continuum of possibilities between a closed skill approach (pre-deciding) to an open skill approach (having an intention that’s open to change). Daryl made a plan for that final touch (closed skill), but, if he had been wrong, he wouldn’t have been able to change his mind and would have lost the touch. He was, in essence, making an educated guess based on the context of that “do or die” moment that he felt would give him the best chance of success.

This ambiguity is part of fencing. And Soren, himself, even fences specific touches with a closed skill approach because he sees value in doing so. He just acknowledges that there are different trade offs of risk and reward. If both fencers are employing a closed approach, whichever athlete guesses right will look 100% smart and the other 100% stupid even though a high degree of luck is involved in determining the outcome! There is value in decisiveness and courage in key moments. Daryl had good reason to think he would be successful because he scored several times in the earlier stages of the bout on this exact action, suggesting it could work again.

However, the open skill approach is more nuanced and difficult to analyze after the fact, which is why, perhaps, it is under-taught (and sometimes downright ignored) in many lessons. Soren felt that he could accelerate his improvement, by spending more training time honing his ability to better deploy an open skill approach in more scenarios. He calls this “building depth” in his actions and believes that focusing at least some of your training time on open skill development is beneficial for every athlete.

Closed Skill Training is Still Important

While the concept of training to “build depth” might seem contrary to typical technique-oriented training, that is not necessarily the case. Why? Good technique, especially technique that has been trained with the goal of being able to adjust in the middle of actions, puts you in a position where you can more skillfully adapt when a situation changes.

So don’t just go in to your coach tomorrow and say, “I read an article on the internet—no more technique for me, thank you!” (although I’m sure many of us may want to). You still have to focus and train this important aspect of your fencing and spend some time drilling technical movements in a closed skill fashion. The reason for this is if your basic technical movements are not ingrained in your muscle memory, you have no chance of adapting to a changing environment. For example, if your parries are too wide and you realize that your opponent is finishing to another line, it will take you too long to bring your hand back to where it needs to be. If you fall off-balance when you lunge, it will take you longer to react and defend yourself properly if your opponent pulls you short. And we’ve written about footwork before, but if your steps are too big and you can’t change tempo smoothly enough, it will open up a window for your opponent to hit you before you’re able to react optimally.

The key point Soren makes is that as you become more advanced, you must shift a greater proportion of your training away from closed skills (doing premeditated actions) in favor of open skills by integrating your repertoire of actions rather than thinking of them as distinct from each other–for example, training your ability to sense when to hit with a feint-disengage or with a beat, and being able to move between those (and more) options intuitively.

Soren’s Approach and Your Fencing Practice

What does it mean for your fencing? How should you be practicing, then? “I would start with a moment that’s going to occur frequently. I started with hitting with a lunge. Then the coach can do slight variations on that starting action to provide additional openings,” Soren explains. “You start with only a couple things that can happen and then you can build up to more.”

The easiest place to start applying this new knowledge is to look at your fencing for a common situation where you often lose touches. Soren’s first scenario was hitting with a lunge, but it can be any action you choose. It can be as general as wanting to work on your attack, or as specific as waiting too long to riposte and getting hit with a lot of remises. If you’re not sure where to begin, you could even ask a clubmate, “Hey, what do you do to beat me?” (just like Daryl Homer would ask his opponents at tournaments!).

Once you’ve identified the scenario you’re going to work on, first try to think of two common things your opponent can do in that situation. When I think about attacking, for example, the first two things that come to my mind are that my opponent could counterattack or parry. Or, if I’m trying to riposte, my opponent could remise or take a counter-parry. There you go! That’s your first drill. Just grab your coach or a teammate, setup that situation, and tell them they can only do those two things (you can also return the favor for a teammate by asking them if there is anything they want to work on). And because most fencing actions are extremely fast, Soren recommends starting at half speed to train the mindset for openness before building up to full speed.

Over time you can make all drills harder in one of two ways: faster or more options. If you’re working on your attack, the next step might be that your opponent can beat your blade as well. Then, when you’re able to do that version of the drill successfully, they could add the ability to pull you short, to do fake actions, or to put out point in line. Whatever it is, you can always increase the difficulty until you’re essentially doing your practice in a real bouting situation--which, after all, is what you’re training for!

Importantly, Soren also recommends a target success rate of 75% to 80% instead of 100%. This might seem counter-intuitive because we are all perfectionists to some degree, but we won’t be getting the most out of the drill unless we are operating on the fine line between success and failure. If the drill becomes too easy, our training partner or coach should increase the difficulty until we start to fail a little bit. In this way, our training can incorporate high quality feedback that is translated directly to our fencing.

Here’s a personal example of how this type of training can play out, step by step: I realized that I often fell short in the middle when I attacked fast off the line. To address this, I created a new drill by doing an advance-lunge against a partner who would either just stand still or take one step back. Then, in the next stage, he was allowed to take as many steps back as he wanted. Then, we did the same thing on the en garde lines: he could either go for a simultaneous attack, or go in and try to pull me short. Then, I gave him the ability to take parries in the middle, do defense actions going back, and so on until there was nothing he couldn’t do, just like in a real bout. All along the way, I tried to think about how my technique was affecting my ability to see and extend my attack. Was my first step too big or too fast? Was I off balance? Was I extending my arm too early? And so on. As I continued to repeat the drill, my ability to attack quickly off the line kept improving to the point where I could see and feel a direct result in competitions.

Advance-lunge against a partner:

  • Stage 1: Partner can stand still or take one step back
  • Stage 2: Add option to take multiple steps back
  • Stage 3: Add option to go for a simultaneous attack or try to pull me short
  • Stage 4: Add option to take parries in the middle and do defense actions going back
  • Stage ?: And so on….

Even if you’re trying to work on something very technical, it may still make sense to train like this. Let’s say your parry four is too wide. The traditional way of correcting that would be to just take parry four over and over and over until you do it correctly. But I find that whenever I’m working on a skill like this, no matter how good I can get it in a lesson, as soon as I’m back out on the strip and under real pressure, I often revert back to the same mistake. By building up the difficulty and variety while you practice the technique, you are training yourself to actually do the action correctly within the context of a bout. It is by doing this that you are building depth in an action. You are mastering slight variations created by different scenarios which actually make it more effective during the complexity of a bout than trying to perfect a single idealized way of executing the technique.

The great thing about building depth like this during training is that you can apply it to all aspects of your practice. For example, rather than just doing some set footwork patterns in isolation, you could start by keeping distance with an opponent. Then you could add a lunge when they open. Then they could open from different distances, and force you to do the correct length of attack. Then they could open and then close a line during your attack, and force you to either finish to a different place or abort your attack altogether. Then they could attack you as well, to force you to pull them short. There are no limits on how much variety you can add.


If this all still seems a little abstract to you, remember, there are only four steps involved:

  1. Pick some aspect of your fencing that you want to work on
  2. Think of a few related actions that can happen in a real bouting scenario
  3. Find a partner and create a drill giving him the option to use variations of those actions
  4. As you improve, add more options and increase the speed

How quickly you progress through these options depends on your skill level. The more advanced you are, the more time you should spend working on drills that have optionalities built into them. If you are a beginner, it makes sense just to repeat one particular piece of technique to get a feel for it, and by the end of the lesson add a second option. Over several weeks, you could progress to more options and more speed. As your skills improve, you could start a lesson with just two options at medium speed to warm up, and by the end of it, you and your coach might as well be fencing a real bout. Keep in mind that no matter how advanced you are, you won’t master a skill in a single day. Developing and executing these kinds of learning experiences should be an integral part of your ongoing training routines.

By incorporating Soren’s philosophy into your fencing, you can take any scenario that actually occurs in a bout and train for it at any level of difficulty and variety. If this way of training seems overly complicated, just remember that it mirrors what you’re doing in an actual bout. It doesn’t make sense to train one way but compete another way! You can go out there and make the best plan in the world, but if your opponent does something you don’t expect--you have to be ready, willing and able to change your plan.

Questions & Clinics with Soren

Feel free to leave questions in the comments, and if I can’t answer them myself I will get in touch with Soren to follow up. Soren also teaches a limited number of clinics each year in which he expands upon this training approach. If you are interested in having him run a clinic at your club, please send me an email or get in touch through our contact form.