The new user interface is in preview!

Want to check it out? Click here! (If you don't like it, you can still switch back)

League Forums

Main - Community Help Forum

Re: Analytic Game Planning: Defending the Hitch (and maybe other plays)

By setherick - League Admin
6/20/2020 9:38 pm
Let’s start by getting this out of the way: the FL Hitch is broken. We all know it’s broken; we know how it’s broken; and we are all going to use it anyway because it is broken.

Now, let’s talk about you can defend it. If y’all find this interesting, then I could be persuaded to do posts about two other plays that I have seen being the bane of about every league I’ve been in since coming back to MFN: the HB Flare and the Deep Corners.

It has also given me an excuse to take a weekend dusting off some data analytic skills, and beginning to branch out into R. (Because while I enjoy playing MFN, those of y’all that know me know I’m just in it for the data.)

The Hitch

The Hitch will destroy your team even when you have game planned for it. There is a reason why you see so many teams win despite only controlling the ball for 20 mins and completing < 50% of their passes.

The reason that this happens, and we all know this so there is little point in re-telling it, but here goes, is that the WR2 breaks down the CB when starting the slant route. If the CB reacts poorly, the WR2 is gone. And if there is no safety deep, the WR2 has an open lane all the way down field.

How about some data?

I’ve been using MFN as an excuse to keep my SQL sharp. Since game logs are nicely downloadable as CSV, I have built a SQLite database in which to transform the raw game log into analyzable statistics because I am that kind of nerd.

The below data is taken from me and another owner’s defensive statistics across 4 leagues (only 2 of which we jointly play in). [If you really want to get into the details of the data or the database, PM me.]

What you’re looking at is a plot of median yards given up against the Hitch plotted against the standard deviation for yards gained on a completion. The data was pre-filtered to defensive plays that have given up < 5 median yards per play in at least 10 attempts.

What you want to look for is the lowest median and lowest standard deviation because that means you are giving up the fewest yards per play against the hitch with a less chance that the play is going to break long.



I only really want to call out a few things about the data.

The Flat Zone – I have seen more owners defaulting to this as their Hitch defense because, as you can see in the data, it works. The one thing that you have to be cautious of with the Flat Zone is the high standard deviation. If the WR2 breaks down the coverage, there is no deep safety help, which is going to lead to a big play. The reason the median stays so low on it is because the linebacker help forces a lot of incompletions.

The 46 Heavy – I love this play for reasons being discussed in the USFL forums. It’s a great all around utility play against the 212 and 122 passes, and as you can see, it keeps the Hitch in check as well. The benefit here is that you have two deep safeties for when the WR2 breaks loose, and one of those deep safeties plays shallow enough to come over on help defense.

The Nickel Double WR1 – This is a wildcard. I don’t want to admit this, but I had overlooked this play. And I can’t tell you when I would have been testing it against the Hitch – I suspect it was during a phase of testing different prevent defenses. Anyway, what’s interesting about this play is that it forces a number of passes to the WR1. This is the other sure way to beat the Hitch. I’m going to be looking into it more, but I’m not sure how much to trust the data on this right now.

Wrap this up already

So there y’all go. I hope you have enjoyed a different way to think about game planning the Hitch. I have certainly enjoyed practicing my data skills.

UPDATE: I updated my database, and the 46 dropped to 3.5 med, so I updated the scatterplot.
Last edited at 6/21/2020 5:36 pm

Re: Analytic Game Planning: Defending the Hitch (and maybe other plays)

By Dogboyuk75
6/21/2020 2:36 am
I think it’s great for community that guys share their info and knowledge ,I’m a stats nerd but lack the technical know how to do anything .i find pieces like this interesting and forces you to open your mind to different philosophies

Re: Analytic Game Planning: Defending the Hitch (and maybe other plays)

By setherick - League Admin
6/21/2020 5:56 pm
I was prepping materials on other plays, but I don't have anything ready to go yet. I wanted to back and touch on a point I was making about the differences between the FZ and the 46 Heavy 2 Deep against the Hitch above.

I noted that the FZ had a higher standard deviation on plays. It's hard to see why when you're just looking at game logs, but it becomes pretty clear when you look at a density plot of of the two defenses.



What you're looking at here is the probability of the result of the Hitch against the two defenses based on the current data in my database. Here are some things to point out:

1) The FZ has a higher chance of forcing an incompletion, so that's why it's 0 point is so much higher. [Here's the deal: nothing sacks the Hitch. Nothing. Seriously, I have 1 sack in my database against it and it was from the Dime Double WR1. The clean % of the play is > 99% in my database. So if you know how to sack it - in fact, if you know how to put pressure on it - I would love to know.]

2) The chance of your opponent completing a 10 yard pass against the 46 Heavy is higher than against the FZ too. That's because the FZ floods the passing lane on the WR2's side.

3) By the time we get to the chance for a 20 yard gain, the two plays are equal. And the chance of plays greater than 20 yards goes hand down to the 46 Heavy. In the current dataset, there has been no plays against the 46 Heavy longer than 31 yards compared to 6 plays over 30 and 4 over 50 yards against the FZ.

Re: Analytic Game Planning: Defending the Hitch (and maybe other plays)

By JiaFeng
6/22/2020 7:24 pm
setherick wrote:
I was prepping materials on other plays, but I don't have anything ready to go yet. I wanted to back and touch on a point I was making about the differences between the FZ and the 46 Heavy 2 Deep against the Hitch above.

I noted that the FZ had a higher standard deviation on plays. It's hard to see why when you're just looking at game logs, but it becomes pretty clear when you look at a density plot of of the two defenses.



What you're looking at here is the probability of the result of the Hitch against the two defenses based on the current data in my database. Here are some things to point out:

1) The FZ has a higher chance of forcing an incompletion, so that's why it's 0 point is so much higher. [Here's the deal: nothing sacks the Hitch. Nothing. Seriously, I have 1 sack in my database against it and it was from the Dime Double WR1. The clean % of the play is > 99% in my database. So if you know how to sack it - in fact, if you know how to put pressure on it - I would love to know.]

2) The chance of your opponent completing a 10 yard pass against the 46 Heavy is higher than against the FZ too. That's because the FZ floods the passing lane on the WR2's side.

3) By the time we get to the chance for a 20 yard gain, the two plays are equal. And the chance of plays greater than 20 yards goes hand down to the 46 Heavy. In the current dataset, there has been no plays against the 46 Heavy longer than 31 yards compared to 6 plays over 30 and 4 over 50 yards against the FZ.

Re: Analytic Game Planning: Defending the Hitch (and maybe other plays)

By JiaFeng
6/22/2020 7:24 pm
setherick wrote:
I was prepping materials on other plays, but I don't have anything ready to go yet. I wanted to back and touch on a point I was making about the differences between the FZ and the 46 Heavy 2 Deep against the Hitch above.

I noted that the FZ had a higher standard deviation on plays. It's hard to see why when you're just looking at game logs, but it becomes pretty clear when you look at a density plot of of the two defenses.



What you're looking at here is the probability of the result of the Hitch against the two defenses based on the current data in my database. Here are some things to point out:

1) The FZ has a higher chance of forcing an incompletion, so that's why it's 0 point is so much higher. [Here's the deal: nothing sacks the Hitch. Nothing. Seriously, I have 1 sack in my database against it and it was from the Dime Double WR1. The clean % of the play is > 99% in my database. So if you know how to sack it - in fact, if you know how to put pressure on it - I would love to know.]

2) The chance of your opponent completing a 10 yard pass against the 46 Heavy is higher than against the FZ too. That's because the FZ floods the passing lane on the WR2's side.

3) By the time we get to the chance for a 20 yard gain, the two plays are equal. And the chance of plays greater than 20 yards goes hand down to the 46 Heavy. In the current dataset, there has been no plays against the 46 Heavy longer than 31 yards compared to 6 plays over 30 and 4 over 50 yards against the FZ.

Re: Analytic Game Planning: Defending the Hitch (and maybe other plays)

By JiaFeng
6/22/2020 7:24 pm
setherick wrote:
I was prepping materials on other plays, but I don't have anything ready to go yet. I wanted to back and touch on a point I was making about the differences between the FZ and the 46 Heavy 2 Deep against the Hitch above.

I noted that the FZ had a higher standard deviation on plays. It's hard to see why when you're just looking at game logs, but it becomes pretty clear when you look at a density plot of of the two defenses.



What you're looking at here is the probability of the result of the Hitch against the two defenses based on the current data in my database. Here are some things to point out:

1) The FZ has a higher chance of forcing an incompletion, so that's why it's 0 point is so much higher. [Here's the deal: nothing sacks the Hitch. Nothing. Seriously, I have 1 sack in my database against it and it was from the Dime Double WR1. The clean % of the play is > 99% in my database. So if you know how to sack it - in fact, if you know how to put pressure on it - I would love to know.]

2) The chance of your opponent completing a 10 yard pass against the 46 Heavy is higher than against the FZ too. That's because the FZ floods the passing lane on the WR2's side.

3) By the time we get to the chance for a 20 yard gain, the two plays are equal. And the chance of plays greater than 20 yards goes hand down to the 46 Heavy. In the current dataset, there has been no plays against the 46 Heavy longer than 31 yards compared to 6 plays over 30 and 4 over 50 yards against the FZ.

Re: Analytic Game Planning: Defending the Hitch (and maybe other plays)

By setherick - League Admin
6/24/2020 9:05 pm
Defending the 221 Long Passes

I thought I would post a few more of these, but if no one is interested, feel free to tell me to stop. As I mentioned, I use MFN anymore as a way of practicing SQL and, now, R to keep my data analysis skills sharp.

Let’s talk about defending the 221 Long Passes – TE Deep Corner, TE Deep Post, and Streak Heavy. These plays have become big favorites of offensive game planners because they all pull the coverage deep and then send the RB on a wheel route in space.

Against most defenses, the RB gets matched up against a slower LB or, even, DE. This gives the offense the advantage because the RB is going to get to space faster and be faster once in it. This gives way to big play after big play.

So what defense should you use? The Goal Line Attack of course. I am personally a big fan of the Attack #1 since it puts a man on the RB, but the other person that I share the database with is a big fan of the Attack #3 because the jailbreak forces the QB to make a decision early.

The data is comparable for both:



What you’re seeing here is that the Attack #1 has a higher probability of forcing an incompletion or getting a sack (those are counting as 0 yards right now because I haven’t cleaned up the db), but also has a higher probability of giving up some yards if the QB gets free from the start.

This goes back to the jailbreak nature of the Attack #3. The jailbreak of the Attack #3 forces the QB to make immediate decisions before the RB gets into position. This limits the big gains because the RB hasn’t gotten all the way into space yet. As the data shows, it becomes more of a gamble than the Attack #1 when you get to the 20 yard mark.

Wait, how do you make it work?

The key to using the Goal Line attack against the 221 Long Passes is overriding two key positions.

1) The RB Cover Guy – If you are using the Attack #1 or Attack #2, you want to make sure that you override the defender that is covering the RB with your best defensive back. This guy should be 90+ SP and M2M ideally. This is the more obvious override so I suspect y’all are doing it already.

2) The MLB Blitzer – All of the attack plays line the MLB next to the “Deep Safety” in the GL set and have the MLB blitz. The problem here is that most MLBs are going to be too slow to effectively get to the QB every time, so you’ll get to watch the QB dump the ball off to the RB without feeling too much pressure. The key here is to find a 95+ SP DB with good tackle. That’s it. That’s really all you need. Run Defense is good for those few times that that DB is going to get eaten at the line on a running play, but most of the time, the DB is going to fly through a gap and disrupt the pass or track down a running back before they can get to the LOS.

This is the most consistent way that I have found to disrupt the 221 passes without having to worry about overuse because you overrode the Goal Line sets with a normal blitz play or coverage play. The additional benefit of using this method is that it is not as vulnerable to the run as running a Dime defense against the 221, which I have seen become more popular recently.


Last edited at 6/24/2020 9:06 pm

Re: Analytic Game Planning: Defending the Hitch (and maybe other plays)

By Dogboyuk75
6/25/2020 4:38 am
appreciate you taking the time to share you info seth,and give your insight into strategies.id consider myself solid roster builder and average gameplanner,but it seems obvious to me to take that next step,overrides are massive part of success

Re: Analytic Game Planning: Defending the Hitch (and maybe other plays)

By Lamba
6/25/2020 6:43 am
I read this, enjoyed this and then 5 minutes later, I'm not sure if he told me that I'm an imbecile somewhere.

@Seth: much appreciated!

Re: Analytic Game Planning: Defending the Hitch (and maybe other plays)

By Phareux
7/01/2020 12:59 pm
I'm sure. lol