[Statistiques avancées] Qui a le mieux réussi sa Draft NFL 2022 ?

[Statistiques avancées] Qui a le mieux réussi sa Draft NFL 2022 ?

Advertisements

Advertisements

Every year experts, journalists and aficionados spend hours deciphering, analyzing and quantifying the performance of future draftees. Once the draft has passed, notes are given, the choices are discussed and the running-backs taken in the first two rounds are mocked.

In all these analyses, it has not escaped the knowledgeable fan that the ratings given as a result of this selection process are as diverse as they are varied. How can we evaluate a draft when its effects will only be quantifiable in a few years? Let’s try to answer this question and apply our answer to Draft 2022.

What is a successful draft?

In a previous article, we explained that it was impossible to draft well over time. The draft process involving too much randomness, it is difficult to be consistent from one year to another. Here it is a question of quantifying the success of a draft without having seen the players on a professional field. Thus the success of a draft at the end of the 7th round can only be analyzed as a consistency of the process of each team during the 3 days of selection. What is valued is not to take a player who will be good in 5 years, because we don’t know (for the moment) anything, but to take a player to the place where he must be taken.

Every pick (or “pick”) has a value, and using one on a player who was available 32 picks later would be a huge waste of resources. A good draft is therefore a draft where the team does not “reach” in addition to taking players corresponding to its needs (the “reach” is the action of taking a player too early, as opposed to “stealing” ).

But how to determine the choice where each player should have been taken?

The Big Board Consensus

Since 2014, journalist Arif Hasan has brought together the “mock-drafts” of 82 NFL analysts. From these simulations, he created a draft averaging these 82 opinions. If we compare this draft that he calls the “Consensus Big Board” to the choices that are actually made on the evening of the draft for each player, we obtain the graph below:

The theoretical choices of the Big Board Consensus are 70% correlated to reality. This similarity is all the more impressive because analysts have far less information than NFL scouting teams. The further we go towards the end of the draft, the more the correlation decreases. This regression is logical, the number of players who can be taken at the end of the draft increases because the level is more and more homogeneous.

Now that we have the location of the draft where each player should have been taken and the actual picks this year, we can look at the gap between the two. A “steal” will be illustrated by a positive value and a “reach” by a negative value.

This method has a dead end. If a team chooses in 5th position a player who should have been taken in 15th position, his “reach” has a value equivalent to a reach from 195th place to 185th. The two cases cannot be treated in the same way. The closer we get to the end of the draft, the greater the number of players who can be taken at these positions and therefore the gaps are less and less representative. You have to weigh the choices.

The Fitzgerald-Spielberger Charter

For this we will use the Fitzgerald-Spielberger chart. In the previous article on the draft we had already talked about this type of charter which tries to quantify the value of each choice of draft and we had proposed an alternative.

Jimmy Johnson’s most widely used charter was seen as overly optimistic and overvalued the early laps. Here the Fitzgerald-Spielberger chart is based on the average amount of contracts signed after the rookie contract to give a value to each choice.

The Fitzgerald-Spielberger chart considers the different choices more homogeneously. Whereas that of Johnson gives choice 31 a value almost 6 times lower than choice number 1. This approach makes it possible to have a modeling closer to reality by integrating the part of randomness inherent in the draft.

Now that we have our method for weighting each choice, we can analyze this 2022 draft.
For each choice we will look at the difference between the value (according to the chart) of the choice where the player should have been taken subtracted from the value of the choice where the player is taken.

Unsurprisingly the Patriots are in last position with many reachs, including the one on Cole Stranger in the first round. The Ravens, who let the draft come to them, are largely first.

We can also look at the 5 best and 5 worst picks of the 2022 draft:

The quarterbacks being projected much higher before the draft, it is not surprising to find two here in the “steal”.

Similarly, it is no surprise that we find high picks because the gaps are much larger at the top of the draft according to our model.

If we exclude the two quarterbacks, the top 5 is completed by Jamaree Slayer and Tariq Woolen.

But why is reach such a bad choice? Ben Baldwin, the reference in NFL statistical analysis for The Athletic media explains it simply:

“On the one hand, the team chooses a player before many others that analysts believe are better. On the other hand, they show too much confidence in their own evaluation skills. Reacher for a player is to send the message that you think you are a better assessor than your peers. Based on previous drafts, there’s little reason for a team to think that’s the case. »

Indeed we had concluded in our previous article on the draft that it was difficult for a team to be smarter than the others, otherwise the system would not be balanced. So no one can afford to play smarter during the draft. The other teams are often as competent as each other.

Of course, we should not blindly follow the consensus of the draft, each team has its needs. Taken on a case-by-case basis, a reach is not always a bad thing, just like a steal can be a bad decision. But by wanting to take players too early, teams are shooting themselves in the foot by lowering the value of their choice. Because even if the draft has a fairly large share of randomness, it follows a certain trend. She is, on average, consistent enough to get the right players out in the right places

There is no guarantee that the players taken by the Ravens this year will be better than the others. But in terms of process, they are the ones who walk away with the most consistent draft crown.

In conclusion, journalist Sheil Kapadia, speaking to NFL teams, said:

“Don’t be overconfident in your ability to assess talent. »

Leave a Comment

Your email address will not be published.