Takeaways: Bill O’Brien publicly gives up play-calling duties at combine presser

The biggest splash of Bill O’Brien’s rare offseason media availability was when he announced just before his press conference to a group of local reporters that Tim Kelly would be taking over play-calling.

This was actually the subject of one of the first pieces on the site — I wasn’t a big fan of the Tim Kelly hiring to begin with. At this point the Texans are handing over play-calling duties to somebody who has never done it before, and I think largely how you feel about this depends on your initial beliefs when Kelly was hired. I’m keeping an open mind about what happens here, but I think as we acknowledge that Kelly has no real background as a coach other than this step, he’s likely going to stay fairly lockstep with O’Brien.

O’Brien followed up his words about this being a big opportunity for Kelly by saying “I don’t think it changes it too much.”

Obviously, it’s basically impossible for an outsider to know how this is going to go just yet. It’s possible that Tim Kelly is the next Joe Brady and the guy who called all the best plays of last season — we wouldn’t know it, there’s no history to draw on. I’m hanging my expectations more on him just being an extension of O’Brien. I did think that O’Brien’s comments about spending more time in the actual moment of the game are a good thing, though.

O’Brien emphasized man-to-man ability on cornerbacks

It’s no secret that the Texans are lacking at the cornerback position and there’s not much settled at all there outside of Gareon Conley. Bradley Roby is a free agent, Vernon Hargreaves was released off the fifth-year option. Lonnie Johnson has to earn playing time.

But this does give us a bit of a compass as far as how the team plans to run things defensively — I think this is a turn towards the aggressive. I think the writing is on the wall that Johnathan Joseph’s time in Houston is done.

O’Brien also emphasized that new defensive coordinator Anthony Weaver would bring “a lot of different ideas and creativity to our defense.”

Watch out, Carlos Hyde

O’Brien fielded a question about running backs in the draft and didn’t even mention Carlos Hyde in the response. I think that tells you about all you need to know about how tenuous Hyde’s grasp on the job will be next year even if he does return.

S**t-eating response on Jack Easterby’s role:

“Hopefully, everybody gets to meet him some day.” šŸ˜‰

You know you’re backpedaling from the start when you’re mentioning the official definition of the guy’s job. It’s hard to see any way around Easterby being a key figure in the Houston organization at this point.

Notably absent things:

The rest of O’Brien’s presser largely bounced around long-term topics like whether 17 games is good, the XFL, long-term deals for Watson and Tunsil, the CBA negotiations, etc. Here are the things I was surprised we got no questions on:

— Nothing on edge rushers at all. No talk about Jadeveon Clowney, no talk about a replacement third guy. No talk about spending an early draft pick on one. I think this team might just be content with who they have at the position.

— No talk on specific impending free agents. I’m sure O’Brien would have mostly rebuffed them, but D.J. Reader and Bradley Roby are pretty big names and I think we could have at least read into his words about whether they’re in Houston’s plans. Darren Fells and Hyde also weren’t mentioned at all.

— No talk on Will Fuller. Fuller’s health is much more important to me than the health of Tytus Howard as far as what it means to this team, and in a draft class that is stacked with wideout talent and some very public comments of frustration about Fuller’s health, I’m surprised nobody pushed on that.


Iā€™m happily writing this article free of charge ā€” this is presented to you ad-free and without any hassle. If you enjoy my work and want to encourage me to produce more, please feel free to leave me a PayPal tip.

8 Things I Learned from Accidentally Becoming a Game Designer

In 2015, after being let go by Bleacher Report, I wanted to take some time and get further into a hobby. I’ve always been video-game inclined, and I had taken to watching Awesome Games Done Quick and branched out from there to watching various RPG streams and YouTubes. My favorite game growing up was Final Fantasy 4 (2 if you’ve only played the US version), and it just so happened that two weeks after my contract was up, an entry-level tournament for the game was about to start.

I finished fourth in that tournament. I immediately became hooked to the real categories, and by 2017 I had the world record in both noCreditWarp (the most popularly-run glitched category) and no64 (the most popularly-run full game category). In 2017, I got to run no64 for Harvey Relief Done Quick, and in 2018, I got to run noCW at AGDQ. I re-routed no64 to create a new, faster endgame strategy that now has all three of the top three times on the leaderboard. After GDQ was over, I figured that I’d start learning new games, content with how it had all gone.

Instead, three weeks later, I was a game designer.


Having been really inspired by Link to the Past randomizer, around April 2017 I started blurting out a bunch of ideas into the #romhacks section of the Final Fantasy IV discord. It was kind of contentious because at the time, what I wanted didn’t exist, and I can’t code my way out of a paper bag. So I went through the trouble of writing down, in very detailed chunks, how an open-world Final Fantasy IV randomizer would go.

— There’d be key items, and those key items would be shuffled between locations where you receive a key item in vanilla FF4. So the Baron Key that you use to open Baron Castle would still be in the game, but you might find it after you defend Fabul, or after you complete the Dark Elf Cave.
— The characters would also be shuffled. Instead of Cecil, maybe you’ll start with Edward. Or Palom. And you could find new characters at the vanilla spots where you’d find characters, but of course they’d also be shuffled.
— Your ultimate goal was to find the Crystal that transforms Zeromus into his final form, which enables you to actually fight the final boss. That was a hard block on progress. You’d also need access to the moon. In the vanilla game, you get that by trading in the Darkness Crystal at Mysidia for the Big Whale. I also advocated for a re-branding of the “Pass” item from vanilla, which just takes you to a largely-useless dance area. That pass item would now take you directly to the Zeromus fight, enabling you to be able to fight Zeromus at different stages of the game.
— You would start with the airship and be able to access anything you had the proper credentials to get to. When you got an item that let you go to the underworld, you’d get to go to the underworld.

I expected nothing to come of this, particularly because the person who eventually created the randomizer typed out in the discord something to the effect of: “boy, I wouldn’t want to be the guy that created that.”

A couple of Fridays after AGDQ 2018, b0ardface launched the very first version of Free Enterprise, an open-world FF4 randomizer that congealed a lot of my ideas and also added a few novel touches which I’ll get into in a bit. Not only was the game a hit, it also created a whole new wave of interest in RPG randomizers. Since it has come out, open-world randomizers have been released for Final Fantasy 6, Chrono Trigger, Final Fantasy 5, and others. I can now log on to Twitch and see someone playing a game I drew up in Google Docs almost any time I want.

By default, I assumed a role that’s sort of difficult to describe: I’m a developer, I’m a community manager who has run six separate tournaments for our game since 2018, I’m in charge of game design, I’m in charge of our YouTube channel, and I’m in charge of submitting our game to various marathons. I hate to make a comparison that sounds so dramatic, but I’m sort of the Godfather of our project — I’ve got my hands on a little bit of everything, and the buck largely stops with me.


This has largely been trial-by-fire for me. It was never my goal to become a community manager, nor was it something I envisioned would happen when I posted some ideas in a forum. There are still times where I’d rather not stream and just play video games on my own so they can be relaxing rather than efforts at marketing myself or engaging everyone’s questions. And this shy person is “in charge” of a 3,000-person community.

Here is what I have learned along the way:

1 — It’s important that feedback and ideas have a place where they can be light-hearted, and it’s important that as a developer you engage in that conversation

There are quite a few people who stroll into our Discord’s #feedback_and_ideas forum and get pissed off about how many memes are posted in there. I have even made some curmudgeonly jokes about it. The truth is that the fact that people are comfortable sharing memes is an incredibly positive sign for our community because it means they aren’t being shamed or dismissed.

I try my damndest to be as engaged in the conversation as I am because if I am not being as open as I can with developer intent, I can’t expect people to be as earnest in their requests. And sometimes, the best requests come from that openness, or at least in reaction to that openness. There have been times where I’ve been too grumpy because it’s something that I can’t really comment on honestly without making myself look bad or otherwise creating a toxic conversation, but otherwise I try to answer every question I can in that forum.

One thing that bugs me about a lot of the conversation about games today is that I can give feedback, but that there’s not really any sense that the feedback was accepted. It goes into a Google form, or some other person agrees with you on a forum and nothing happens. There’s no conversation. It’s one-sided, and maybe if you’re lucky, your thing gets created. It was what I originally felt when I came up with the concepts for Free Enterprise: my ideas weren’t really being heard by someone who could implement them. The community I’m about is one where we answer as many questions our audience has as we can — even if those answers aren’t always what they want to hear.

2 — You must empower the correct people

Over the course of my time running the community, I have added about 10-15 different people as workshop helpers. I’ve added even more people as community testers and discord moderators. Some of these people have grown the game beyond what I could have even expected — they’ve created tournament concepts, they’ve created racing bots for the discord, they’ve created alternate color palettes for the characters. They’ve created scripts for magic damage, and they’ve created sites that log race results. They happily organize restreams for the game. We’ve got tutorials for learning how to restream, how to commentate effectively, and so on.

People who are willing to help often show it through their actions. My job is to read that, intention and give them that power. It’s very easy to get caught up in just what you want to do for the community — I largely want to make game play changes and modes, but if I hyper-focused on that, everything would get stagnant. I’d be in a communit-me instead of a community.

Another thing I want to focus on that I could not have done on my own is that we’ve created a very LGBTQ+-friendly, inclusive community. That didn’t happen because of me — it happened because I empowered people who have been excluded or shamed for this in the past, and they made it a point to integrate an inclusive lifestyle into our community. If you asked me how much inclusivity mattered to me before this, I would have told you it was important, but including gender tags on our restreams wouldn’t have been one of my ten most-important things to fix on our scene from my cis-het, game-obsessed point of view. By empowering the right people, they became a priority.

When I think back about giving people power who have actually demonstrated that they want it constructively, I have never been disappointed. It’s a big deal for the community that people like that are properly rewarded. Thus, it’s become a big part of my job to be on the lookout for people that add to this and empower them.

3 — You have to pick a lane and accept that the lane isn’t going to be everyone’s favorite

Free Enterprise is meant to appeal to racing — racing is the foundation of the randomizer. It certainly helped that the predecessor romhacks for FF4 tended to appeal more to straight-forward challenge plays than a brisk race atmosphere, in so much as it became a major differentiator for us.

But we still get plenty of ideas coming in to the feedback about ways to make the randomizer more challenging and lengthy, or ways to make the randomizer more wacky. Some of those we keep, some of them, like the concept of a Playable Golbez, become memes that I will never escape.

I’m very polite when I say no, I’ve had to learn to say no a lot since we’ve started this. I’ve also had to learn how to say “maybe later” — because there are a lot of things that are cool ideas, but not necessarily important on the development path for us in the near-future, with our one-person development team. It’s okay that you enjoy other games, and it’s certainly valid for people to want Free Enterprise to be more things to them than it currently is — but it’s also my job to keep it on the path that it’s supposed to be on.

4 — If you’re creating a new tool or approach, ask what you’re breaking and if that is a good thing

When we were very naive, b0ard and I had an idea to put text hints into the randomizer. You would talk to the NPCs, and NPCs, as video game NPCs would sometimes do in RPGs, would drop hints. On its face, when you say it like that, it’s a clever turn on traditional RPG play.

What happened when we implemented it is that the backlash was so hard that we removed them from the game within two days, convinced it would never work in original format.

What happened? We forgot that we were a racing game. When you race, you’re accustomed to actually playing the race game. The original hints were so powerful that people would spend 15 minutes reading NPC text, know exactly what to do to beat the game, and be annoyed. We had replaced the racing game with speed-reading.

I’m not going to promise we’ll never make a mistake again — I’m not even going to promise we won’t bring hints back again in a different form — but that was a context that I hadn’t thought about before we created it, and we paid the price for it.

5 — If you explain enough to newcomers to get them asking questions, then you can bring them over the rest of the way

The most important thing I ever did for the randomizer was create a newbie guide. The newbie guide has drawn some criticism over the years from some community members because it doesn’t tell people exactly what to do. In fact, there’s been clamoring for a more advanced guide for years, but nobody has ever really done it beyond small sectioned chunks for specific concepts.

Here’s the reason why the newbie guide worked: It didn’t tell anyone what to do. It gave some recommendations, and it explained how things could be different in different scenarios. It was a launching point for learning things. And that launching point got people asking questions to the community at large, and that drew people into the community. To date, our #newbies_corner is one of the most popular channels in the history of the discord. We have people who have developed videos of strats in relation to frequent discussions about things in the corner.

Most of the best runners of the game start with these questions the newbie guide sparked, and they then gradually learn more and more. Other than myself and penguin8r, most of the best runners of the game aren’t the best runners of the vanilla game. Becoming great is something akin to an accumulation of great knowledge rather than a script that you learn. If you look at the guide purely as an informative text about every scenario you can face in the game, critics are right, it’s incomplete. But if you look at the guide as something that is a gateway to learning more, it’s actually been pretty successful.

6 — Whimsy and fun matter as much — if not more — than game design

When I talk about designing the game, I talk about the actual structure of the game. The two most important things we have in the game are randomized whimsy that generate hype:

— Because Zeromus is hard to look at, we have a flag that replaces him with one of 400+ other bosses, pop culture manifestos, and so on. Maybe you’ll face Bowser, or maybe you’ll face one-winged Barkley from Barkley Shut Up And Jam: Gaiden, or maybe you’ll fight Poochie.

— Because the TwinHarp track is kind of dull, we improve it by randomizing it with one of many, many songs from other video games as played in harp form. You could get a Chrono Trigger theme! You could get the Game of Thrones theme! There are a lot of different options on the table.

Between those two things, we generate a ton of hype and excitement over otherwise benign events from a casual perspective. To that, I owe a lot of credit to b0ardface — nowhere in my original design did I even account for this idea, but it is something that unites viewers and runners alike with hype. And, of course, we both owe a lot of credit to SchalaKitty and Calmlamity for their work on making the sprite and song pools continue to grow.

I don’t know if I’ve ever written this out anywhere, but I think one of the reasons that b0ardface and I were able to create a great randomizer is that we come from entirely different perspectives and enhanced different areas of the game. It’s not my original vision. It’s our original vision. It was kind of the perfect storm of everything coming together, not something that’s easy to emulate.

7 — I’ve had to sacrifice my ego in a lot of ways for this to be where it is today

The general way Twitch works for 95% of people is something like: I’m going to stream, five people will watch it maybe, and that’ll be that. I live in a zone somewhat beyond that, because I’ve gone to some big places and done some big things. But I’ve had to really adjust my expectations on what kind of payoff I’m expecting to get from this in a few ways as well, which is an interesting concept to deal with when you created something that gives joy to a lot of other people.

— All of the Free Enterprise channel’s bits and sub money goes to our head developer, b0ardface. This is in part because if we started divvying out slim pickings, a lot of people actually have some space to claim them. There are a ton of people helping out, and the interest tends to wax and wane. Also, as much as it is my idea, b0ard does most of the work of the randomizer itself and it feels right that he gets the main reward.

My own Twitch channel, mostly through a lack of activity, has exactly one non-me subscriber. A lot of the time that I’d spend playing video games instead gets turned into me designating changes, or responding to feedback, or otherwise handling “the business” of the randomizer. That in and on itself can feel like a full-time job. It is my belief that not being able to stream more has cost me marathon appearances this year, not to mention not even applying for some places because my brain wouldn’t even go there.

— I don’t really treat the game the way I would if I were just someone who ran it. I’m not looking for tiny optimizations here or there because if I found them, I’d have to fix them. So, I’m not on the cutting edge of strategy as a general rule. I honestly think it’s best for the randomizer if I don’t win any of the leagues or tourneys it puts on, not because it’s a conflict of interest, but because I feel like if I win I didn’t teach anybody else enough about how to play the game well. As someone who is top level at the vanilla game, this is a really interesting scenario to find yourself in — you feel like chum. You feel like you’re helping other people make big names off of you.

— Other people have stepped up in the absence of that and made names for themselves, as I expected. So then I have to deal with the fact that I could get that tiny bit of recognition, and I could be the person who gets the small rewards that it brings — the respect and recognition, the bits and cheers. But instead I’m … working on adjusting prices! Spending that time answering questions! Thinking about which ideas come next! It’s not very glamorous.

So, while I appreciate the people who have recognized me with a donation or their respect — in a general sense, I might as well be completely off the map of a game that wouldn’t exist without me. It is … an interesting place to be. I can’t tell you it doesn’t make me happy to see other people having success because of what I put out there, but I also can’t tell you I’m thrilled that I don’t really share in it much. It’s a mental blockade that I struggle with a lot.

8 — People care about fairness so, so, so much

You all do. I’m sorry, you just do.

The two biggest controversies that we’ve dealt with as a community have come because of the idea of “what is fair?” In our first Highway to the Zemus Zone league, our initial tournament document dared runners to use the 64-floor glitch because b0ard had created a very clever punishment for anyone who would dare try such a thing the traditional way. Instead, a runner took the 64-floor glitch to abuse the logic of the game and skip needing one of the items to get to the final boss.

In our second league, Highway 2 The Zemus Zone, multiple players were accused of stream sniping because of suspicious plays that could, potentially, have been influenced by what they saw on the screens of others. We had no hard, concrete proof, only things we could make inferences from.

When I changed the randomizer logic to make the Earth Crystal treasury yield higher-end items, people were incredibly emotional about it.

In each case, what mattered wasn’t what happened — but whether what happened was “fair” — fairness is the word that launches a million conversations and five-paragraph discord screeds. What’s more, your definition of fairness may differ wildly from mine and my subsidiaries. There’s a reason baseball fans can’t click on a website without being inundated with Astros cheating stuff. We as people take stock of what we believe is right and just and fight vociferously when other people’s thoughts don’t match ours. That is the spirit that launched further with the relative anonymity of the internet, and as a developer, I have to plan for it. I can’t let loopholes be created because people will climb through them. Nothing can derail an otherwise hype event more than a day-long, 300-message argument about what is fair and what is not.

When people start telling you that something is unfair, you have to listen. It doesn’t mean they’re automatically right — it does mean that you have struck a real nerve somewhere and figuring out how you did that is important.


Iā€™m happily writing this article free of charge ā€” this is presented to you ad-free and without any hassle. If you enjoy my work and want to encourage me to produce more, please feel free to leave me a PayPal tip.

A Statistical Look at Potential Texans Veteran Pass Rush Options

I’m going to start this with a belief: I don’t think the Texans are going to make a move for any actual pass rush in free agency. I think when they signed Whitney Mercilus they valued him as a core member of the team for the next two years, and I think the obvious solution is to draft somebody in the second or third round, then let that pick, Jacob Martin, and Charles Omenihu fight to see who takes over Mercilus’ role by 2021.

But let’s approach this from a blank slate. Here’s how our pass rush splits look for Mercilus, J.J. Watt, Martin, and Omenihu over the last three years:

If you pull up a list of free agents, it’s easy to fawn over the top guys. I don’t think the top guys are going to actually make it to real free agency and that leaves me with some tough decisions about who to feature versus who to not feature. My belief is that Shaq Barrett will be franchised and that he’s a lock to stay in Tampa Bay. My belief is that Yannick Ngakoue is going to be franchised as well, though I think that situation could get ugly enough to where Ngakoue gets traded. That’s an eye-raiser to me — Bill O’Brien has shown no interest in making his draft picks and I think he’d consider Ngakoue a building block ala Laremy Tunsil. The question is simply if the Texans are willing to pay what Ngakoue thinks his market is. I’d speculate you could probably get that done if you parted with your second-round pick, and maybe a bit more.

A-List, No. 1 guys who are probably not coming here:

We all know why Jadeveon Clowney isn’t coming back. That horse is so dead that White Walkers would have trouble re-assembling the bones to re-animate it. Barrett is 27, Clowney 27, and Ngakoue is just 25.

Fool’s Gold

These two players both worry me because I think they fall into blind spots for Bill O’Brien: First-round pick pedigree has tended to be a big deal for BOB, and each of them is coming off a career year. Lawson is the more preferable of the two options to me, but I have a lot of respect for Sean McDermott and would expect Lawson to have more trouble in Anthony Weaver’s scheme. Dupree’s defense popped off as soon as Minkah Fitzpatrick joined, and it led to a sack explosion for him. In the long view, these are both solid second banana rushers — but they’re going to be paid like No. 1 players in my view.

My favorite fits

Bold – I used his Jaguars stats instead of his full-season stats, not counting mid-season change against him

Judon is a tough evaluation. On pure numbers, he’s up there with the best of the class, but Baltimore’s heavy-pressure system also means he’s not lining up and beating his man over and over again on all of those pressures. I think it’s fairly likely he draws a tag, but I wanted to mention him just in case.

Fowler and Armstead are funny to me because Armstead stays extremely close to Fowler on a statistical level and still plays inside. Because of his ability to satisfy a 3-4 end position, I think Armstead probably bumps up a notch on the Texans own preference chart, and I think that makes him an ideal fit. Armstead was also already playing well before the 49ers defense became dominant, which I think helps forecast him to being a solid fit in Houston’s defense.

Fowler is coming off the fabled 10-sack season that Jadeveon Clowney never got, but I can’t imagine he’ll come cheap enough for that to matter to the Texans. Pass rushers get paid, and the second Mercilus inked that extension, I think it ruled the Texans out of a player like this. But I love the fit, and so he’s a favorite fit.

Old players and the Texans under Bill O’Brien

Every single player that the Texans have signed as a real-dollar free agent since 2015 (or traded for, even) has been under 30. The closest to being over 30 was Matt Kalil, who at 29 was an extremely desperate move to a position of need. Vince Wilfork happened in 2015.

I happen to think that aging players are a great way to supplement the core at edge rusher — Robert Quinn is coming off a terrific season in which he led the NFL in ESPN’s pass-rush win rate. I was quite intrigued by Bill Barnwell mentioning in his look at the NFC East that Ryan Kerrigan might be traded for — I think he’d be an a-plus get for the Texans, even coming off a season where he missed four games.

But obviously I’m not going to project the Texans to wind up with anyone in this age bracket because they haven’t gone there in five years, and there’s very little happening in the front office that should make you think they’ll suddenly be changing their ways on that.

Value players I could see the Texans taking a chance on

Vic Beasley is 27 and has 37.5 sacks in his last four seasons, including a 7.9% hurry rate and a 10.1% pressure rate. Those numbers stack up pretty well with most impact free agents, yet he’s got the stink because he wasn’t a 1A rusher in Atlanta like they drafted him to be. I think he’s a good buy-low candidate.

Jordan Jenkins is just 25, but notched 15 sacks over the last two seasons despite starting only 23 games. With a career pressure rate of 10.4% and a career hurry rate of 7.5%, he’s another guy who I think is getting a bit overlooked because he’s coming from a bad situation. He’s only a former third-round pick though, so will O’Brien even dignify his presence?


Iā€™m happily writing this article free of charge ā€” this is presented to you ad-free and without any hassle. If you enjoy my work and want to encourage me to produce more, please feel free to leave me a PayPal tip.

The Carlos Hyde Conundrum

Carlos Hyde had, by all normal respects, a traditionally successful year for the Houston Texans. He rushed for 1,000 yards for the first time in in his NFL career, averaged 4.4 yards per carry despite a career-high number of carries, and played exactly 50% of the offensive snaps. He was a big part of Houston’s offensive identity in 2019, and he started off hot.

The problem was that, after the hot start, Hyde didn’t really do much to elevate the offense. He’s a stellar back with great vision, he’s got the ability to run out of arm tackles, but he doesn’t have breakaway speed and he’s not going to evade people in the open field. When he was used as part of a Deshaun Watson-focused game plan where teams had to respect Watson as a runner, he excelled. When he was asked to plod into the line with inside zone, he was usually dispelled. From the bye week on, when Hyde ran inside zone per Sports Info Solutions, he gained 3.1 yards per carry on 52 attempts. Just one of those attempts gained more than 10 yards. He also fumbled four times:

RB table courtesy Football Outsiders

Only 20 NFL running backs reached 200 carries last year, and I imagine if we were ranking them out on pure talent, Hyde going on 30 years old probably isn’t in the top 15. I’d say he’s better than Sony Michel, who has been a bust for the Patriots. I’d say he’s better than David Montgomery to this point in Montgomery’s career. You can probably give me a few other “I dunno, (Le’Veon Bell) has been slipping for a while…” players as well. But generally speaking, guys who are the focal point of your offense are supposed to be better than Carlos Hyde. You give 200 carries to Saquon Barkley or Aaron Jones. I think Hyde is a better fit as a goal-line back or in a less-impactful role than he had last year — and I’m saying this as someone who has plenty of respect for Hyde’s overall game.


Hyde wants to come back to the Texans and the Texans shouldn’t be particularly bothered by meeting Hyde’s demands. The running back market has tanked, and the teams that are willing to pay for one aren’t going to pay for one with as much mileage as Hyde has. So, practically speaking, I think I’ve chalked this one up as a keep for the Texans. There’s not a logical reason to replace Hyde unless you can get someone clearly better, and the only player hitting free agency who I think is clearly better than Hyde in a Hyde-shaped role (early downs, goal-line, run-heavy sets) is Derrick Henry. I could see an argument for Melvin Gordon, though I think a lot of his value comes from the fact that he gets used in the passing game.

The shadow of Duke Johnson looms over Houston’s running game in an interesting way. Johnson is clearly a fantastic tackle-breaking back, the Texans have committed to him for the long-term by dealing a third-round pick for him. Johnson has been successful at every opportunity despite block-reading not being a big strength for him. But at the same time, no coaching staff so far has been comfortable letting Johnson be the lead dog of a committee. Bill O’Brien didn’t even scratch the surface of what Johnson could do in this offense in my opinion.

Therein lies the innate problem: It might not be logical to want someone better than Carlos Hyde — speaking purely on an analytical level, it would be downright idiotic to spend limited resources that the defense needs on that. But if next year’s offense comes back with no changes, one of the best ways to improve the offense is to find a player that’s more explosive than Hyde to take those carries, be it Henry, a high-round pick in an NFL draft that has devalued stud backs, or someone else. If you look at the other positions on this offense, everything is more or less locked in by money or draft position except right guard, tight end, and starting running back.


There’s a certain part of me that believes that the Texans are so far in on Bill O’Brien that they might as well just let it all ride. And that side of me thinks that if the Texans are going to run their offense like they did last year, they might as well pony up a competitive offer to Derrick Henry. Running back is a market where a lot of teams are playing scared, so you might get a bit of a discount, and you’d also be robbing the Titans of their bread-and-butter offensive player at the same time. If you want to run into eight-man boxes, get the guy who ran into them 112 times and still averaged 4.35 yards per attempt (per SIS) rather than this guy:

As dumb and pointless as I think quite a lot of O’Brien’s early-down runs are, they become a little more understandable when you put someone like Henry in the backfield. At that point, they no longer become sustaining offense, but offer home-run elements in their own ways. Of course, I say this and Duke Johnson had a higher broken tackle rate than Henry did last year. But unless O’Brien is willing to rethink how he uses his backs, I can think of very few things that would upgrade this offense more than putting the ball in an actual elite back’s hands.


When I think of Carlos Hyde’s 2019 season, I think of found money. He was traded over here at last cuts and performed incredibly well at times. After watching Alfred Blue play running back for five years, anybody who can read a blocking scheme is a hero in my eyes. The only negative in my eyes was taking Arian Foster’s jersey number, and that’s not something he had ultimate control over. I also want to be clear that nothing I have heard so far has made me think that the Texans don’t plan on re-signing Hyde.

But I do think if the Texans sign in to 230 more carries from Hyde next year, they’re going to regret the decision to make him a focal point of the offense. He’s just not that level of player, and he can’t handle that kind of snap count without his lack of flexibility hurting you in a meaningful way.

Ultimately, that’s more a function of how O’Brien uses his backs more than it is about Hyde. But that war is long and gone, and we have to approach this offseason with the idea that you have to build everything around O’Brien. If GM O’Brien doesn’t idiot-proof running the ball for coach O’Brien, our plural O’Brienmen might be less than pleased by the results.


Iā€™m happily writing this article free of charge ā€” this is presented to you ad-free and without any hassle. If you enjoy my work and want to encourage me to produce more, please feel free to leave me aĀ PayPal tip.