What Ten Years of Teaching Krav Maga Actually Changed

I've been teaching Krav Maga for ten years now, and if you'd asked me last month what had changed in that time, I'd have given you a pretty confident answer. I learned and incorporated more disciplines. I developed more perspective and more seniority. And I've helped to develop an exceptional new generation of coaches. All of that is true. But it's not the more nuanced story that the data tells.

Thinking about this topic, I fed Claude approximately 1,500 of my actual lesson plans going back to 2016. I asked it to find the real pattern of growth and development in my data, instead of confirming the one already in my head. What came back corrected me on more than one point. Here's what I learned.

Krav Maga is built to be simple. I stopped needing it to be.

IMO, Krav Maga's design philosophy is one answer per problem, wherever possible. One technique for a straight punch, one for a front choke, taught pretty much the same way to everyone in the room. If you're a soldier with six weeks of training or a cop who drills four times a year, you need something that works under stress without a decision tree attached to it, and Krav is among the best in the world at doing that job.

My earliest lesson plans look exactly like that. A 2016 plan for defending a low roundhouse kick reads like a highly focused checklist: warm-up, power drills, demo, practice, pressure test. Every single technique got its own, specific lesson plan. There was just one approach, one answer, and I taught it to everyone in the room.

But here’s the reality: Forge students aren't training for just six intense weeks. A lot of them are in the room three times a week for a decade, and they show up with different bodies, different backgrounds (so many Taekwondo black belts!), different instincts under pressure. I suspect this is true for many American, civilian krav maga schools. Somewhere around 2022, my plans start solving the same attack across multiple levels in the same class. By this 2026, techniques show up in my lesson plans with several different but viable answers, in multiple modalities. A Krav Maga double leg takedown starts with the fundamental two-hands-from-the-front version. But now includes a modern arm-drag to takedown version I picked up from Ryan Hoover, and two different BJJ answers, side by side in the same lesson plan.

I don't think the original Krav answer is wrong. I think it was built for a population that isn't training at Forge San Francisco in 2026. Once you have the time and the range of students who train regularly and want to go deeper, staying at one answer starts costing you something important.

I started writing down whose idea it was

Nothing in my 2016 or 2017 plans has a source attached to it. A technique just was what it was. Usually lifted directly from the Krav Maga Global playbook. The first time that my notes cite outside material by name is 2018, and it's a small thing, a statistic pulled from a knife-attack study. In 2026 everything is named and date-stamped and usually with an accompanying URL or video. This is Ryan Hoover's read. This is what I saw Jared Wighoni teach at the PTTA conference. That behavior has only become routine in the last couple of years. It's definitely not a slow decade-long habit.

I'm still working out exactly why this mattered enough for me to start implementing it. Part of it is a simple, grateful case of offering credit where credit is due. There are so many practitioners, teachers, thinkers, explorers and innovators out there and they deserve to be recognized in whatever humble way I can offer. Part of it is so I can remember it more easily, give it a point of reference outside of a flat curriculum document. I can't tell you how many times I've gone back to reference a third party video from my notes before teaching a class and incorporating it.

All that said, old material didn't get deleted as Forge evolved. It got labeled. Current version, KMG version, KMW version, Paulo Rubio version. Eli Knight version (thanks for the excellent standing RNC defense, Eli- we built it into our curriculum). I love that my historical records put each of them sitting right next to each other, so I can choose between them as necessary and on purpose. Looking back at my 2016 lesson plans, there was one version of most techniques. Now there are several, clearly documented, and picking between them is part of the coaching job (and part of the fun).

I got specific about the how, not just the what

We used to teach reverse hammer fist, blade of the hand, blade of the forearm as four different techniques. More specifically, we just named them all on a couple of lines in our curriculum documentation. No methodology for teaching. No unified way for students to practice them. They're practiced and demonstrated in one flow now. As range diminishes, the strike moves up the arm, fingers to the hand to the elbow. It's the same motion adjusted for distance and intent. But learn the flow once and you can practice it with all the various striking surfaces. It's a much better way to teach and learn the ideas, I think (credit to Jared Wihongi's Pekiti University for that).

Third-party protection followed the same path at a bigger scale. It used to be a single vague mention in our curriculum. Now it's a progression that spans four levels; protect, escort, intervene, integrate, and a student or a coach can point to exactly where they are in that journey. Everyone at every level knows how they are supposed to implement it and it's written that way in our curriculum - with clarity and specificity.

The plans themselves got more precise too. Early ones are full of placeholders, "technique TBD," checkpoints reduced to a fragment like "chin down, shoulders up." A 2026 plan reads like something I could hand to a coach who's never taught the class before: contact points named, coaching instructions spelled out, the order of the sequence written down. The curriculum stopped living in my head as a loose set of concepts and became much clearer and more detailed as we revised it year after year.

Scan and escape stopped being the whole answer

Krav Maga's roots are strike and get out. Hit hard, create the gap, leave. It’s what we call “scan and escape.” That's genuinely good advice in a lot of self-defense situations, and it's still core to what I teach.

But the world my students train for has changed. MMA and BJJ went mainstream over the last decade, and most people walking in now have already seen grappling and control work as real tools, not just theory. In plenty of the situations closest to home, mine included, striking and leaving isn't always the right answer. When you have to protect a loved one, control becomes a real need, not a replacement for hit and run. It's a choice made based on context. As Ryan Hoover puts it so elegantly “It depends.”

That's part of why the Forge graduate curriculum is richer now than it was five years ago, more BJJ, more Kali, more options. It isn't padding. It's giving students the control option that Krav alone never gave them, and the kickboxing and knife boxing combos added at that level are the practical version of the same idea. Different approaches for different situations. It's giving a larger self-defense toolbox to a craftsman who can use the tools effectively.

You can't be a complete defender specializing in one thing

Someone who only strikes has a gap. Someone who only grapples has a gap. Someone who only handles weapons has a gap. My historical lesson plans show my own gap closing over time, and it's something I can actually point to rather than just assert.

There's a stretch in my historical lesson plan record where Kali shows up constantly, an intense period tied to a specific stretch of my own training, and then it gets integrated into the ongoing material without needing a spotlight every time it appears. That's roughly what integration looks like when it's real, absorbed rather than added as a novelty or dismissed out of ignorance. BJJ picks shows up in my lesson plans about a year later and by 2025 gets cited even more often than Kali does. Ground work stopped being its own island around then and started getting the same ranked, sourced, integrated treatment as everything else in the curriculum.

Striking, grappling, weapons. You need real capability in all three to call yourself a complete self-defense practitioner and protect what you care about.

What didn't change

Not everything changed. My punch mechanics, hip-as-power-generator, snappy recoil requirements. They read almost word for word the same in 2016 as they do in my lesson plans today. The basic shape of a class, warm-up, conditioning, technique, a closing pressure testing element, hasn't moved in ten years either.

P1 - our foundational material - belongs on this list too. The earliest Forge classes a new student takes look a lot like they did a decade ago, and I think that's appropriate. That material has to work for someone with zero background walking in for the first time.

More curriculum, and a different person teaching it

The Forge curriculum is richer and deeper today than it was ten years ago, especially at the graduate levels, and I'm not going to pretend otherwise. That part of my original personal, internal narrative held up fine.

What changed more than the curriculum is me. Ten years ago I was a new teacher, running the class in front of me. Now I'd call myself a coach, with a wider set of tools and a better sense of how to read a specific problem and a specific student before reaching for an answer.