Z.03: Bringing Data to a Gun Fight
Part of the reason we’re still failing to make a data literacy pivot in the army is because we keep delegating that responsibility down to non-commanders. Data is a leader’s responsibility, not just some staff officer’s or data nerd’s job.
But too often, even when we do get leaders talking about data, it often only focuses on admin functions. Certainly not something like lethality. ‘Lethality’ and ‘readiness’ are two words we see used a lot in the military, without a lot of data to back them up.
But two recent examples show the possibilities when leaders leverage data to do exactly that. First, the USMC’s new JMAP leverages data to transform the way we train and measure close range combat lethality.1 In the army meanwhile, we have 4th Infantry Division running ‘Moneyball for gunnery’ to improve armored crew success in gunnery while driving down the costs to achieve qualified crews.
Shooting is Fundamental
The changes in Quantico actually began back in February 2018, when then-Secretary of Defense, James Mattis, directed the creation of the Close Combat Lethality Task Force. It was stood up to give the ground combat soldiers, where nearly 90% of military combat casualties occur, the same focus we have given expensive platforms like fighter jets and aircraft carriers. The Marine Corps Rifle Marksmanship Lethality Capabilities Based Assessment (CBA) was one such effort.
One of the first things the CBA identified was ‘…efforts to assess and provide improvements to marksmanship lethality would be temporary without a new way to quantify and assess enterprise-level marksmanship lethality.’
‘Without data, the benefits of this CBA will be short-lived and the idea of continuously adapting to a thinking enemy will once again be relegated to anecdotal assertions rather than quantifiable capabilities.’ - USMC CBA, 2018
God bless the USMC, they took the CBA to heart.2 They set out to quantify and measure lethality, and discovered it’s not just accuracy, its accuracy and speed at scale. With JMAP, they shifted away from the old points mentality to one of lethality. It’s fixing the old ‘just counting things that can be counted’ paradigm to one that quantifies a SPEAR model — speed, precision, executive control, adaptability, risk exposure.
The old rule ‘If you can’t group well, you can’t shoot well’ still applies. But now there is ‘accountability per round’. I’ve lived what the authors of JMAP Stewart describe: ‘I don’t know which ones were good, which ones were bad, I just see a target with a bunch of holes in it.’ With JMAP ‘…now we have accountability and a time associated with the shot taken … They can read the data and the data will tell them what to do.’
One particularly insightful change was to drop the ‘maximum score’ model of old. The Quantico team discovered the points model wasn’t able to reflect the changes they saw in the civilian shooting world. ‘Last decades grandmaster would get smoked by today’s. Looking at our own records we don’t have that improvement. All we have is a plateau’. ‘We didn’t have the right datasets… we had a ceiling effect on our marksmanship.’
‘I am not currently incentivized to do more than my maximum amount of pull-ups’ one of the article’s authors noted. Changing to JMAP means focusing on data driven and measured lethality, not ‘I got expert’. Now, ‘It’s accuracy with time’. There is no max score. ‘Someone can always get a little bit better’. The authors see this as a key fundamental change. There’s lessons here for the army, to include with our ACFT, if we want to learn them. ‘Competition breeds excellence’, and they’d like to see a portal where the entire USMC can share scores and challenge each other.
‘Moneyball for gun crews’
Meanwhile, over in Fort Carson, Colorado, another commander found a way to leverage data the army already had to change the way their armored crews validated their lethality in gunnery.
It all started with a commander’s question. Lt. Col. Jonathan Bate, who leads the 2nd Battalion, 23rd Infantry Regiment in 1st Stryker Brigade Combat Team asked ‘We’ve got all this gunnery data… but do we know what's correlated with better success?’
‘A gunnery crew must pass six types of tests, or tables, to earn its marksmanship certifications: a written test, a simulation, in-person range training with lasers and blanks, two live-ammo sessions, and the final qualification round. These semi-annual and annual certifications are required across multiple weapons platforms, such as tanks, Strykers and Bradley ground vehicles. For Strykers, there’s a crew of three—a driver, a gunner, and a vehicle commander, but only the gunner has to qualify on the table exercises.’
When they poked at the data, Bate found that, ‘…out of all the tables, there was one table that turned out to be statistically significant, correlated with achieving a first-time qual’. And it wasn’t the one anyone anticipated. Instead it was the laser-and-blank range.
‘It was really unexpected to us,’ Bate said, expecting the live-fire shooting to be more impactful. But shooting is more than just trigger time. It’s also about fighting as a crew. ‘Use the right fire commands, like, “troops in the open”...and using the right terminology before you engage a target, and acquiring the target.
“If you score below, let's say like an 800 or 850 out of 1,000 on that table, you go back and retrain, because you only have a 70 percent chance of achieving a first-time qual. So it saves us time. It saves us ammunition and resources, and helps us overall become more successful,” he said.
4th Infantry Division’s change to gunnery was built on ‘very smart lieutenants doing the data analytics’ paired with ‘a really smart sergeant with the common sense—the experience’. ‘We put that all together, and he helped us interpret the results from the data,’ Bate said.
Leaders Make the Future Happen
‘Never in our Corps’ history has a Marine had access to such quantifiable, combat-focused lethality data’. With JMAP, trainers can ‘…take this data and compare the time and accuracy of each shooter to determine what would be the most effective use of the unit’s training time on the range or in dry fire.’
And that data revolution to shooting doesn’t stop at the marksmanship range. The USMC leaders argue they have a ‘running experiment’ which they want to expand to other weapons like the machine guns, the Mk19, and the Carl Gustav.3
This data driven approach can also be leveraged to change the way we acquire new systems.
‘Because JMAP enables a data-centric approach to decision making, the Service no longer needs to make assumptions as to whether a new weapon, holster or any other individual equipment or training curriculum has a performance advantage relative to another. We can instead give the new equipment, weapon, or training to a group of Marines and rapidly know quantitatively whether it produces positive or negative results. In other words, JMAP facilitates decision makers to have instant access to real-time performance data to make data-driven decisions.’
Back out in Colorado, Bate wants to build an automated tool using machine learning to ‘take all of these factors and then pop out a predicted score on table six,’ which would flag crews that need retraining.
‘The tools are out there. It's really just a fight for the data, for the will to use it,’ he said.
Striking Out in Softball
In contrast to the leaders in Fort Carson and in Quantico, we have the example set by a senior army leader presenting at the Army War College last month. Addressing a class of rising colonels, the visiting general was asked by a member of the audience, ‘'What does data literacy mean to you, and what do we as future leaders need to prioritize for data literacy from the tactical to operational level?'
His rambling response was a bit of a marshmallow fruit salad, never once cohering into anything which leaders should prioritize:
'Let me flip that on you. You guys, I'm a digital immigrant... So what does data literacy mean to anybody in the room? What does data literacy mean to you? What does it mean when someone says ‘Hey, you need to be data literate’. I mean it's ones and zeros, right? And it's… it's formatted. It's constructed. It's unstructured. It's structured.'
‘Digital immigration’ shouldn’t be an excuse. Data literacy isn’t new, nor is its application to the military and war. Florence Nightingale — yes, that Florence Nightingale — was looking at the data from the Crimean War for combat lessons back in 1856.4 The general went on:
It's not that you got to have a PhD or a master's. But do you know enough of it so that when CD [Lieutenant General Donohue] wants to do a warfighter using live data and the sim collapses on itself because it can't handle the volume of live data that's going out there. CD is data literate. He's data literate enough to go back in and say start asking the right questions.'
Lieutenant General Donohue is a great example of a trilingual leader and was cited as such by Mikolic-Torreira and Emeila Probasco in the podcast I cited last week. But the guest speaker didn’t seem inclined to follow his junior’s example.
'So, I'm never going to get a PhD or a master's in data, but I've tried to educate myself to the point where I know how to ask the right questions…. so that I guess that's how I'd answer that question.'
It takes a lot of cognitive dissonance to have no idea what data literacy is, nor see it as a leader responsibility when prior to the question-and-answer portion, his slide deck said data transformation was a leader responsibility, applying to every one of the Chief of Staff of the Army’s four priority areas.5
The army is trying to pivot to ‘network-centric warfare’, and right there at the foundation of the pyramid on his slide is data.
But this general was presenting from the ‘I have a staff for this’ and ‘I haven’t needed it thus far’ camps I wrote about previously.
'How do we infuse people in our formations that help us with all this stuff? How many chief technical officers do you need in a division? How many chief data officers do you need in a division? How many should be wearing the uniform? How many should be civilians? How many should be contractors? Should it be an MOS? Should be a profession? Should it be a functional area? I don't think we've kind of figured that out yet.'
The general offered ‘Martec’s Law’ as his excuse for why we aren’t demanding more from our leaders. This one was new for me, so I had to do some googling. Far from a law, Martec’s post focuses on how the speed of tech outpaces the speed of organizational change. It is not supposed to be an excuse to do nothing. Martec’s argument is clear, ‘Great technology management is choosing which changes to absorb**’.** That comes from leaders and leadership.
The leaders at Quantico and Fort Carson are not CDOs or CTOs. They are commanders. They are driving change with data. Meanwhile our senior-most leaders are still missing ‘what is data literacy?’ softballs, despite having a slide deck full of data demands. Right now, the most impactful way the army is driving the change to Network-Centric Warfare and data literacy is via retirements.
Joint Marksmanship Assessment Package. The marines of Weapons Training Battalion in Quantico recently published an article in the Marine Corps Gazette, and did a podcast interview with War on the Rocks, which this post cites heavily. You should give them a read / listen.
As I’m writing this the day after, a happy belated birthday to the USMC.
The Mk-19 is a belt-fed grenade launcher, and the M3 Multi-Role Anti-Armor Anti-Personnel Weapon System (MAAWS) is a recoilless-rifle affectionally known as ‘The goose’.
Hat tip to Natalie Stone for enlightening me about her work.
Warfighting, delivering ready combat formations, continuous transformation, strengthening the army profession.