Picking progressive passers

July 03, 2019

I've been looking more at progressive passes and what it might tell us about spotting players of high potential early on in their career.

Original article here

Progressive passes, as defined by Wyscout, are forward passes that are 30m long when the pass starts in the team's own half or at least 10m in length in the opponent's half.

So far I've only looked at the volume of progressive passing. This has been because I cannot filter the data on the player ranking section of Wyscout other than by age. If I choose per 90 statistics a lot of players with a tiny number of minutes show up and push other players off the results screen (limited to 30).

However, my suspicion has been that by using the volume of passes I am effectively looking at young players who played a lot. Young players who play a lot of games in lower divisions will get picked up by bigger teams. So maybe progressive passing isn't the great stat that I keep saying it is?

How about I look instead at the accuracy of progressive passing and restrict it to midfielders and forwards?

I'll look for any player with over 1000 minutes play who is classed as a midfielder or attacker and appears in the top 20 list for players in their league listed by progressive pass accuracy who are now <24 years old

I also only looked at France in my first attempt. How about Italy and Spain?

ITALY SERIE B 2015/16

1. Luca Torreira

2. Frank Keissé

3. Alexis Zapata

4. Stefano Sensi

5. Rolando Madragora

6. Luca Mazzitelli

7. Mattia Aramu

8. Moses Odjer

9. Bruno Petkovic

10. Nicolo Barella

A very good start. Of the players listed only Zapata and Aramu have dropped down a level. Odjer has remained in Serie B and everyone else has moved up to top tier or international football.

But is the problem still there that I am restricting it by age too much and therefore just showing all the players who were young and playing lots?

How about looking at Spain and increasing the age range to players 25 and below

SPAIN SEGUNDA 2016/17

1. Enzo Capilla

2. Santi Comensana

3. Inigo Ruiz de Galaretta

4. José Pozo

5. Amath Diédhiou

6. Gonzalo Melero

7. Iban Salvador

8. Joan Jordan

9. Damia Sabater

10. Sasa Zdjelar

11. Nando

12. Ager Aketxe

Less successful. The main difference is that there are more older players appearing in the list. In the case of Joan Jordan this has picked up another excellent player. Also we have far more wide attackers appearing rather than deeper midfielders.

This could be showing that stylistic differences between leagues have an impact with more space out wide in second tier Spain compared with second tier Italy?

So maybe in Spain we should forget about accuracy and look for players in central areas who still attempt to progress the ball, and try the per 90 stat but manually filter the results.

So same year, same restrictions we'll look at per 90 progressive passes from midfielders and attackers.

1. Fabian Ruiz

2. Joan Jordan

3. Sasa Zdjelar

4. Damia Sabater

Only 4 players make the list that is dominated by defenders with low amounts of minutes. However top of the list by a mile was a young midfielder called Fabian Ruiz currently valued around £50m. The top 2 were miles ahead in terms of progressive passes per 90 and minutes played.

So what about the last 2 seasons of data? I can filter that by minutes and progressive passes per 90.

First of all the test data for a league I know. For all players with over 1500 mins ordered by Progressive Passes per 90 and who play in midfield or attack.

Now I want to restrict the minutes so that anyone who has played at all appears.

Shelvey and De Bruyne have now made the list.

And restrict to 23 or under:

Some new names now including Bacuna, Pereria, Anguissa, Quina and Winks.

At the bottom of the list are players you might expected such as Calvert-Lewin and Solanke who are less involved in build up play.

So if we apply this* to the Segunda and Serie B what names come up as good, young, progressive passers?

and

* had to slightly fiddle with the age/mins settings as the data was picking up too many tiny minute samples but these are roughly the same as the Prem data.

And for a test lets chose, at random Switzerland and Romania, although we can already see that data-driven Brighton have signed the number one progressor in the Romanian league.