Sunday, April 06, 2014

Draft Strategy

So today we'll take a break from sifting through the players themselves to take a look at the actual strategy of drafting. Because the draft is a complicated thing. Getting it right can set a team up for years to come. Getting it wrong can set a team back for years to come.

There's a good reason why teams that consistently draft well such as the Giants, Packers, Patriots, Ravens and Steelers have remained competitive for more than a decade, and why teams that tend to draft poorly such as the Rams, Browns, Jaguars, Redskins and Texans are sitting at the bottom of the pile.

But what does a "good" draft look like versus a "bad" draft, and how do you go about building one?
In his book "Finding the Winning Edge", Bill Walsh made the point that if only around 50% of your draft picks go on to become regular contributors to your team then you've done well. Drafting players and projecting their success into the NFL is actually incredibly difficult, for a number of reasons, and as such even if just 3-4 of your seven picks go on to be useful players then that's not too shabby at all.

Compounding the problem is the nature of the draft. 

People have a tendency to look at it just as a matter of picking a bunch of players off of a list, like picking teams back when you were in high school. But the financial element of drafting adds yet another dimension to this. A first round pick is not just a player, he's an investment. Even with the rookie wage scale in place a first round pick is still going to cost you around $5 million per year with all his bonuses included, and will only get more expensive as time passes.

Thus getting that first pick right is not just a question of your teams on field production for the next few years, it's also a matter of balancing the books and extracting maximum value from each selection. For this reason we start to introduce risk profiles into what is an already complicated and uncertain process.

Before we get too deep into that though, let's look at the two predominant strategies used for drafting; Best Player Available (BPA) vs Needs Based Drafting (NBD).

BPA is simple; your pick comes up, you look at your draft board, and then you pick the guy that's at the top of it. He's the best player available at the time, so you pick him, regardless of what position he plays at. 

NBD is a little more complicated and requires you to take into account the state of your own team. If your pick comes up and the best player on your board plays at a position that you're already pretty well set at then you keep going down your board until you find a player who plays at a position that is weak on your team. So if you have a decent quarterback on your team already and the best player on your board is a quarterback then you skip him and move on to say wide receiver, where perhaps you have no decent players (like the Panthers this year).

The reasoning behind the two strategies is very different. Picking the BPA is a long term strategy with the goal of building up your team over time to keep them perennially competitive. Drafting using the NBD strategy is designed to fill gaps in your team quickly, generally with the understanding that you will quickly turn your franchise around and/or make a push for the Super Bowl in the near future. Almost every organisation claims that they draft using the BPA strategy, but looking at their picks clearly indicates that many teams don't.

So which is better? Fans across the league bemoan each year their team taking yet another cornerback when what they really need is a linebacker, but the reality is that teams that genuinely pick the BPA tend to have the most success in the long run. I used to think along the lines of NBD, till I sat down a few years ago and really thought about it. Then the merits of the BPA strategy became obvious.

The best way to think about this is to look at some of the amusing examples that people come up with to defend the needs based strategy. Why - for example - draft a quarterback in the first round when you already have a good quarterback on your roster? I mean, if you had someone like Brett Favre on your roster, would you really go out and draft someone like Aaron Rodgers?

Erm, yes, is the answer to that. That's precisely what the Green Bay Packers did and it worked to perfection. Favre was getting old and Rodgers was too good to pass up. Three years later Favre was moved on and Rodgers, having had time to develop, was introduced as the new starter. Two years later he won a Super Bowl and now he consistently leads his team to winning seasons and the playoffs.

That's because drafting has to be more than a "win this year" philosophy. The best teams are the teams that draft with an eye to the future. And by the future I don't just mean two to three years, I mean five to ten years. Rodgers was an investment for the long run, the definition of a franchise quarterback. It didn't matter that Favre was the starting quarterback at the time. What mattered was what the team would do post-Favre. Rodgers was too good to pass up and drafting him essentially set the team up for sustained success over the next decade.

It's the same reason that the Giants persist in drafting pass rushers and wide receivers, or why the Steelers keep drafting defensive front seven guys and offensive linemen. Because they don't care if they're stacked at a position this year. They care if they're stacked at the position for the next ten years. It's why the Patriots drafted offensive tackle Nate Solder a few years back despite having one of the best offensive lines in football at the time.

You also have to consider the value of loading up with high draft picks at certain positions. Think about the Texans this year. They hold the first overall pick and may select South Carolina defensive end Jadeveon Clowney. Some people think that would be a waste because the Texans already have a great defensive lineman in J.J Watt and that instead the Texans should take a quarterback.

But think about it this way for a second; how impressive will the Texans defense be if they have two quality defensive linemen, both capable of rushing the quarterback and both capable of stuffing the run? It's not just a case of improving their defensive line. With two players of that caliber they improve their entire defense

They give quarterbacks less time to throw and running backs less opportunities to get out of the backfield. Their corners and linebackers will have a much easier time of things as a result. And if the defense is playing well, holding teams to low scores and low yardage, then that takes a lot of pressure off the offense as well. Now they don't have to put up 30 points per game just to stay competitive.

So even though the Texans draft a defensive lineman, it helps them at other spots. The team as a whole gets better. Now they have the luxury of picking up a quarterback in the later rounds or as an undrafted free agent and developing him over time. They could just wait till the next draft even and see if a quarterback comes up on top of their board at some point.

Compare to teams that draft poorly, reaching for position players instead of taking the top guy at the time. Most often this transpires with quarterbacks, as teams allow themselves to be convinced that just because they badly need a quarterback then that justifies drafting one in the first round. This is how you end up drafting guys like Blaine Gabbert, setting your franchise back three or more years and tying up big chunks of cap space on players who then under perform, while everyone else around you gets better drafting the best players available.

Now let's be clear, if it's a close run thing between two players and the team is struggling to decide who is better, then at that point you could argue that one player being at a position of need could tip the scales in their favour. I could live with that, just as long as we are talking about two players who genuinely are very close in terms of talent.

The other strategy I've seen proposed that a team might adopt would be a hybrid approach, whereby the team drafts the best player available for the first few rounds, then comes back to a position of need later on. That sounds fine in theory, but when you consider that by the time you get to the third round you're already getting into territory whereby the bulk of the players will not pan out in the long run then I think that's a waste of time. You're just as well off grabbing the best player on the board and then filling the need in free agency.

The next concern for teams is trading. Do you trade away picks and jump back into the first round to get that key player you have your eye on, or do you perhaps let a top player slip away by trading backwards and accruing more picks to build with? 

This is a subject that is easier to talk about in theory than it is to do in practice. The reality is that all of us are human and naturally we all suffer from nerves. We might be looking at a certain player who is considered a third round pick by most, but we convince ourselves that someone is going to take him at the back end of the second, so we get in their first by trading up for that mid-second round pick.

And of course if you take a guy earlier than he's projected then you're immediately branded as stupid in the media because you "reached". This is the part of the draft that really gets on my nerves the most. 

Everyone in the media has their opinion. Everyone sitting at home has their opinion. And everyone is entitled to their opinion. But nobody is entitled to criticise a team for "reaching" until they find out how the pick turned out. You can argue that in your opinion it was a bad move, but you can't accuse someone of reaching just because you disagree with their valuation of a player.

As I stated back in my 2012 quarterback draft analysis I would have happily taken Russell Wilson with a first round pick. At the time I would have been branded as a moron, having "reached" for Wilson. But right now I'd be sitting on a two time Pro Bowl quarterback and rookie of the year winner. Oh, and a Super Bowl winning quarterback.

Value is relative and in the case of rookie prospects its entirely subjective based on the opinions of the people doing the evaluating. When I watched Blaine Gabbert back when he was entering the draft I saw a valueless player, someone who I wouldn't go near with any pick. Most of the professional drafting community saw a first round pick. Ultimately I was right, but at the time none of us knew that. I suspected it, clearly, otherwise I wouldn't have said what I said about him. But I couldn't really accuse the Jaguars of "reaching" because in their minds he was the guy they wanted. 

So that's my first issue out of the way. Ignore any talking heads who talk about reaching for players. Perhaps if a team drafts a guy in the second round who everyone thought was a seventh rounder then you can raise a quizzical looking eyebrow and wonder why they didn't hold on for another round or two, but just remember that ultimately if he turns out to be a quality football player then the fact that he was drafted in the second round will become irrelevant.

Is it worth trading away a mass of picks for someone though? Hmmm.

I'm not generally convinced. My issue is that there are very few players that come up who you can genuinely say it's worth selling the farm for. It's almost like an admission of defeat, that you have such little faith in your ability to pick good players later on that you're prepared to trade away the chance to select two or three good players in exchange for banking all your hopes on one.

These sorts of trades also have a very poor history of working out the way teams thought. Even if you do snag a good 'un, you sacrifice the chance to select extra players to help build around him and often set the organisation back at least a year. My experience has been that every draft is quite deep, providing you trust your own analysis and not that of most of the TV talking heads. If you're confident in your ability to find solid players later in the draft then I think you're better off holding on to the picks and waiting. Build for the future, not the present.

Finally we get to what I was talking about earlier with risk profiles.

So, question; what the hell am I talking about? Answer; ... well, I'll explain.

There's a reason that an 18 year old kid who's just passed his driving test and bought a sporty hatchback pays more for his car insurance than a 40 year old woman with two kids and twenty years driving experience who's just bought a family car. 

One is considered as having a high risk of being involved in an accident by the insurance company, and one is not. Experience has taught the insurer that regardless of the specifics about who the youngster is, people in his age group, with his experience level, and driving the car of his choice, are more likely to have an accident at some point. As such they demand a greater risk premium (money) in order to hedge against the possibility that he will smash into a tree or another car than they would for the middle aged lady who is an experienced driver and likely to be regularly carrying her kids around with her.

There are no guarantees in life. As Clint Eastwood once said in one of my favourite quotes of all time "if you want a guarantee in life, buy a toaster,". But there are options that are less risky than others, like our middle aged lady driver vs the inexperienced kid. The same applies to draft prospects.

There are players that are more prepared for life in the pros than others. There are some players that are almost immediately ready to hit the field and then there are some players that need a lot of work to develop properly. For me "potential" is not good enough. Every player has potential to a greater or lesser degree. Just because a player may have a "very high ceiling", that does not mean that he is guaranteed to actually climb that high.

And as such when I rate prospects I like to take risk into account. It's the reason that I don't rate Sammy Watkins and Mike Evans as first rounders (which is where both are likely to go). Not because they're not good football players, but because they're risky prospects. While both have excellent physical talents, neither is particularly that skilled right now. Both need significant work on their route running skills to get the most out of their athletic potential and as such both are risks.

They may turn out just fine. They may turn out to be exceptional. But both might equally prove unable to translate their innate athletic performance into something usable at the next level against much better quality opposition, ones who are just as physically gifted and who eat receivers that can only run three or four different route patterns for breakfast. 

Watkins and Evans represent a level of risk that is unacceptable to me for a first round pick. As I said at the start, a first round pick now commands on the order of $4-5 million per year, with an expectation that their salary will rise further when they come to sign their second deal. For that money, which is the equivalent of signing a pretty good free agent, I want a player who I don't have to worry about, one that will almost certainly contribute to the team in a meaningful way over the course of that first contract.

When you consider the quality that's available in the first round each year then there really is no reason to take a massive punt on someone who only has "potential" as a commending feature about them. To go back to the earlier point about teams that draft well, look at how often a team like the Patriots, Packers, Ravens, Giants or Steelers will take a guy with "potential" versus a guy who most people agree is ready to plug in and play right now.

The best teams tend to take sure fire picks, people who they are convinced will start sooner rather than later. Teams that take a lot of flyers in the first round based on the prospects "potential" tend to be the teams that draft early a lot. See the saga of JaMarcus Russell for a perfect example of what happens when you draft players based on what they might one day become based solely on their physical potential.

Now as you move down the draft order this becomes less and less of a problem. This is the risk premium that I was talking about earlier. As you move down the rounds the salary level drops and the likelihood that the player will work out in the long run also drops. If you want to start taking flyers on people who perhaps didn't perform that well in college but who have a good combination of physical traits then this is the place to do it.

If they work out as you'd hoped then brilliant. If not then you can cut them without having to take a massive salary cap hit. This is the place for finding two things; bargain players who slip through the earlier rounds and developmental players that need a lot of help to become great. Here, not the first round or the second round, is the place to start taking risks. 

Imagine if JaMarcus Russell had been a fifth round pick instead of a first. The fact that he's now out of the league wouldn't mean a thing. Nobody would care, not even the Raiders. Why? Because he would have cost them very little and everyone knew he had the potential to be a gunslinger. So he didn't pan out? How many fifth round picks do? Only one of the fifth round picks from the 2007 draft has been to a pro bowl, and only a handful of the others are even still in the league.

By assessing risk properly, and charging a premium to account for that risk, teams are able to help steer themselves away from the players that simply end up as dead cap space and more towards the solid picks that franchises are built around.

If you liked this post then don't forget to share it with your friends, and if you haven't already then check out my posts looking at quarterback, running back and wide receiver prospects in the 2014 draft (links below). Next up will be tight ends, then offensive tackles, interior offensive linemen (guards and centers), defensive ends, defensive tackles, outside linebackers, inside linebackers, corners and finally safeties. 


Running Backs;

Wide Receivers;

No comments: