First – what’s the plan now for you post-XFL?
I’m not certain right now. I’m looking for the right opportunity – I had a tech background before the XFL so I’m trying to decide which direction to go – there are a lot of routes I can go.
I would still love to pursue my legacy of rule changes – I’m trying to see what the best route would be to push those forward at the NCAA or NFL level. But I’m not sure if that’s a near or long term plan.
What would it look like to you – in a perfect scenario?
I want to help the NCAA and NFL understand how to innovate while keeping the essence of the game intact. To enable the rules to create a safer game for the way it’s played and officiated.
The most important thing is the process – the output is what the fans want. But you need systematic change. I worked in Silicon Valley and you need a process. In the XFL we knew it was the fan’s game first. That was how we went into it – Oliver made it clear. So we went into it from that perspective.
How did you go about ensuring that?
We talked to experts and created a data set – including KPI’ – to match the issues fans had with the sport and we tested for those. You build and repeat. We ran a game under the new rules with two junior colleges first and taped it and watched the video and learned from that. We tested the game. It took years to keep testing to get to the right point. We started off with 100 rules and tested 25, and came up with 15 we implemented. We knew what we wanted to be and knew it would change as technology changed. Netflix always wanted to be a streaming company but it didn’t start that way until the technology got there, for example. You have to be willing to adapt.
But we used what technology we had – had play calls go to position groups not just one player for example – that helped us adapt the game.
What were some of the big things you focused on?
Special teams was a unique opportunity for us. The way we did it -we identified the problem first. The safety issues was the run-up on kickoffs at the other team and those collisions – so we focused on reducing those collisions. We didn’t eliminate the play – we eliminated the problem, and ended up with no injuries on kickoffs.
And the extra point – the NFL pushed the extra point back to eliminate give-me plays and reduced successful kicks from 99 to 95 percent. Well, five percent doesn’t get a pat on the back. We kept games close instead and made plays exciting by offering different opportunities for teams to score instead. The problem wasn’t the play- it was the choice. We identified the problems using data.
How do you use data for all of this?
The key is to open up the data to use in different ways. Data needs context. Next Gen Stats measures backs in miles per hour – yet players don’t run for miles. But the stat adds context. You need to contextualize the data.
For example in scouting – we use the data to help identify players to watch, If the data shows something different than what we see on film, then it helps us take a closer look at that player. But you rely on the film at the end of the day.
When you draft a running back too. What is Le’Veon Bell’s true value? He ran behind a great offensive line including my former roommate David DeCastro – James Conner too. How do you analyze that? You don’t need data to show when a great player makes a great play. But you do to determine the value of some things. Like, as a former lineman, the idea of the running game acting like body blows to a defense was big for me. You could see the will evaporate in players’ eyes when you wore them down – see them start yelling at each other and getting upset. But the data showed that that wasn’t real – it didn’t help you win games more. So which do we believe? Does the run game help you win in other ways then?With play action so it matters in different ways? We need to use data to help guide our decision-making.
Bill Belichick too – he has his own way he drafts players. He has a process. He gets as many picks as possible – uses a shotgun approach – and selects as many guys as possible with the same body type and figures if he gets enough, some will work out. That’s his process and tendency.
Well, how do you figure that out then?
Some of it is common sense. Like passing the ball more – the data shows that defenses get called for penalties at higher rates than offenses do on passing plays – and the likelihood of big plays happen more in the passing game. Which leads to increased chances for touchdowns – which is what the game is about. It often comes down to doing what’s not expected though – that’s what helps make success happen. The best runs are on third and longs, for example. So getting back to wearing teams down with the run game. How do you use the power run game more effectively? You don’t have to do that in an I formation. You can do it out of the gun – out of the 11 formation – when it’s less predictable – and that gets the other team’s nosetackle off the field. It’s about keeping your identity but doing it differently so teams don’t know what you are doing. That’s what data helps you do – find different ways to do what you want to do so you can be less predictable. You don’t have to lose your identity to be different.
Ben made that possible in Pittsburgh but he did it on his own through freelancing. They’d run a screen but inexplicably he’d throw it downfield to AB. He breaks all the rules doing that – but it worked because it was unexpected. They let Ben run the show that way.
How else do you see it play out like that?
Well it’s interesting – I played under Roman at Stanford and he built that pistol offense from Nevada’s. Those duo plays work because of the talent they have in Baltimore. But they pass out of that offense to tight ends 80 percent of the time. When you get to the playoffs those types of tendencies can be studied more and stopped. That’s why you have to keep doing what you do but find different ways to do it -data helps show those tendencies.
And on the rules – we wanted to know why we wanted to make changes- not just to do them. We wanted to speed up the game – that was what the fans wanted. One way to do that was to shorten the time between plays, There were three elements to that – calling the play, the huddle, and line of scrimmage.
It took 31 seconds on average to call a play – including 19 to call it, and eight to get to the line of scrimmage.We solved for those in different ways. If it took Aaron Rodgers eight seconds to get everyone to the line of scrimmage we knew it would take our XFL guys at least that long. So we left that alone. So we addressed calling the plays. We took a year of data and realized that the length of the calls were a big issue. We worked with the coaches and changed the length of the calls – made each play call shorter and had them shorten their playbooks. And we added microphones in all of the helmets – except for the offensive linemen – so the playcalls went got all of them at once. That sped up the playcalls.
I’m sure the coaches were very resistant though…how did you get them to agree to that?
Well, we didn’t go into it trying to reduce the playcalls by, say 10 seconds. We worked to solve the problem.
The coaches were very resistant at first. But we talked to them about approaching this as a place of opportunity, not of fear. We wanted to be able to solve the problem and make this the game the fans wanted.
I had a coach that told me right before the first game he was sure we’d have to change the kickoff rules- no way they would work. But it worked and he saw that. We didn’t assume anything – we studied and used data to support the decisions.
I was worried about the three-point play. We had no data models for this. It was a nightmare entering data for something that was never done before. I was scared but most excited about that rule. It was what the fans wanted though and it worked. We listened to them. Henry Ford said ““If I had asked people what they wanted, they would have said faster horses.” We knew we had to listen to the fans, but had to use our own processes and address the actual issues.
Read more by former Steelers via the book Steelers Takeaways: Player Memories Through the Decades. To order, just click on the book:
