Draft Theory: There Are Levels To This Game

The title quote comes from current UFC heavyweight champion Daniel Cormier, one of the greatest MMA fighters of all time. He said it to remind people that there exists a massive gulf between the truly elite and the hot prospects who look great but haven’t faced the top level of competition, even if it’s impossible to see it until they face off.

I don’t think I need to explain how that applies here.

I have written about cognitive biases and fallacies many times in the past, but this isn’t a bias or a fallacy. This is simply a natural flaw in human thinking. Humans excel at creating heuristics to quickly categorize and process a massive volume of information in a reasonable amount of time. In the vast, vast majority of situations, this way of processing information is greatly beneficial. But in the pursuit of expediency, fine precision is lost.

Draft Tiers

It is interesting to see how that manifests in draft analysis. It is generally accepted that a big board has tiers, everybody in a higher tier should be taken over everybody in lower tiers, and that the tiers should remain as objective as possible from year to year.

Zion Williamson is a #1 pick quality prospect. Nobody disputes that. But there are levels to this game. Some #1 overall prospects (Simmons, Davis) are better than other #1 overall prospects (Towns, Irving, Wall) are better than other #1 overall prospects (Ayton, Wiggins) are better than other #1 overall prospects (Bennett, Bargnani). You may not agree with the placement of each guy there, but however you had each guy ranked heading into each draft, the point remains that not all #1 prospects are created equal.

That’s the easy part though. Zion? Morant? There can be quibbles about where they fall on the top prospect scale, but they are top prospects in some form. There’s a big difference between being Steph Curry, Damian Lillard, and Kemba Walker, but if you get any of those guys, you’ve had great success. You have a max franchise player, and you just need to figure out how to build from there.

However, when it comes to draft projection, Tier 1 encompasses all of these guys typically. It’s a huge range, but very few people make the fine distinction between “top 5 player,” “top 15 player,” and “top 30 player.” I am guilty of not doing it. There are huge differences in these categories, yet top prospects are top prospects regardless of just how top they are.

The Challenge After Tier 1

It gets much dicier as you move down the list. Only 0-3 players each year fit that mold of “unquestioned top prospect.” Everybody is going to hit or miss on those guys together. The moment you move down even one rung of the quality ladder, all hell breaks loose. Again, this is a matter of making these “tiers” that are buckets that are too big, inconsistent from year to year, or otherwise just don’t do a good job of accurately reflecting the prospects.

There are really fine distinctions between good, average, mediocre, and bad players. The difference between a good player, average, and bad player could be something easy to spot like 3-5% shooting. But 5% really isn’t much – that’s one in twenty. A 40% 3P shooter needs to display the barest of other skills to see the court. A 35% 3P shooter needs to show many other skills. A 30% 3P shooter needs to be either a point guard or a center, or else they’re unplayable. 3P% is notoriously difficult to project. Why? The difference between great, average, and bad is the difference between 8/20, 7/20, and 6/20.

The difference could be something much more difficult to spot. Off-ball/help defense. On-ball defense. Boxing out or rebounding. Shot release speed. Off-ball movement. Handling or passing. Ability to understand and execute principles. Ability to operate and process at NBA game speed. These are all little things, but they all matter. A shooter who doesn’t understand off-ball movement and doesn’t have a fast release is not a shooter at all, even if he hits a high percentage when he does shoot. A defender who doesn’t react to off-ball movement quick enough isn’t much of a defender at all, even if he is a quality on-ball defender.

There is a major difference between Buddy Hield’s terrible defense and Devin Booker’s non-existent defense. There is a major difference between Al-Farouq Aminu’s inconsistent, below average 3P shooting and Semi Ojeleye’s plain bad 3P shooting. And the problem is that just a 3-5% difference in quality can be the difference between a player being a solid roleplayer and a player being pretty much unplayable.

There are levels to this game

We’re not talking about one made or missed shot every five games. We’re not talking about three or four defensive plays a game. That is the literal reading of 3-5%, but that is not the true effect. An offensive player who can’t shoot is ignored. Yes, his personal production isn’t good enough, but it also affects everybody else’s production because they’re now playing 4 on 5. A player who isn’t good enough defensively isn’t good enough defensively on every play, meaning the help is not there when it should be and extra help needs to be committed when the attack is direct.

For the top level players, these things don’t matter as much. They have top level skills that overwrite the normal rules. Joel Embiid’s defensive fundamentals are severely lacking, but his size and athleticism make that moot. JJ Redick can’t pass, can’t dribble, and can defend at the barest minimum level, but his off-ball movement and shooting speed and accuracy are so high that he just doesn’t really need anything else. For players with elite skills, ignoring their talent because they have some weaknesses is pointless. For other players, every little thing needs to be dissected because it just takes one or two skills to not be up to snuff for the whole package to be unappealing.

Elite Skills

The ultimate result of this is a severe misunderstanding of which prospects are safe and why. The safest prospects are not the ones with the most well-rounded skill sets but the ones with elite skills. If a prospect doesn’t have an elite skill, it just takes one or two things to be just a tiny bit worse than projected for him to go from good to bad to unplayable. If an elite skill is a bit worse, it’s still a great or a good skill. Yes, the other skills have to be a little better, but there’s simply more wiggle room.

Most prospects don’t have any elite skills, and that’s why it’s so difficult to project them. Take a bunch of 18-22 year olds. Predict how they’re going to develop. If you’re 3% too low or 3% too high, you’ve missed badly. 5%? Forget about it. And if a prospect is 3% worse at just one thing, they may be just a fraction of a percent worse than expected, but depending on what it is, that could still be enough to put them on a much lower level. And because of the nature of draft projections, it can be extremely difficult to put prospects in the same context as previous years’ prospects or NBA players.

Efficiency and Reputation

That last part is where a large majority of disagreement comes from. Objectively evaluating something subjective is, well, difficult. Even after a lot of tape study, two people can see entirely different things. No two big boards should ever be the same, and this is the primary reason why. If two people are projecting the same player to improve but are starting from very different points, the end result is going to end up in two very different places.

The players that generate the most disagreement are inefficient scorers. There is a tendency to handwave away obvious glaring flaws with prospects and players who score a lot of points and look good on highlights but who do not score efficiently and do not provide enough other value. Andrew Wiggins is a prime example. Ask of every player, “If he took less shots and he scored less points, would he still be considered a good player?” Unless you are Lillard-level good on offense or better, if the answer to that is no, then there’s major danger. In this year’s draft, RJ Barrett and Romeo Langford stand out as guys who may fail this test at the next level.

The other major category of players that end up with wildly different rankings are those who have a reputation not supported by the scouting reports. If a player is touted as an elite defender, but the scouting report is “While a good defender, he does tend to get smoked by quicker, explosive players—which he’ll see far more of in the NBA” and “Lacks defensive playmaking skill; doesn’t log many explosive plays in the blocks or steals columns,” he’s probably not an elite defender (those are from De’Andre Hunter’s scouting report on The Ringer). Basically, sometimes players get reputations that are not supported by their scouting reports. If the reputation doesn’t match the scouting report, you’re going to end up with wildly different valuations based on which you put more stock into.

And remember, having just a 5% different opinion on a player is wildly different. Good luck!

Previously: On Patience and Probability | Next up: Leftover Food for Thought