Vincent Janssen

[Trilly]

Farmer

 

[razörist]

This is my issue with De Boer, the level of play in Holland is so so weak that even he as a manager will have to earn his stripes elsewhere before getting a shot at Arsenal.

He's made his decision, about staying or leaving, will be public after Ajax won the title. Hope he stays, afraid he goes to Everton because most desirable positions are filled right now...

I kind of agree with you, Arsenal is a big step. But on FDB as a manager, I don't quite think you can value his performance based on the competition. Most of his titles he did not have the best team, and still won it. He does not spend a lot of money, its all about him. For most of the season his first 11 included a fully ajax youth back four, Bazoer who joined when he was 16, Klaassen is an Ajax youth, El Ghazi joined as a 16 year old. That's 7/11 players, and has been quite normal since he took over at Ajax. I think he can do a lot better still, I really do.

 

[BobP]

Finnbogason has like 6 goals in 11 appearances for Augsburg, not too bad.

 

[BobP]

Also, the thing is, some of those players like de Jong and Alfonso Alves moved to clubs where their talents were not utilized effectively.

For example, de Jong moved to a Gladbach side who placed a heavy emphasis on mobile, transition based football. How is a striker like de Jong, 6'2 and over 85KGs, supposed to fit into such a team?

 

[ferrarif50hunt]

21? He looks about 35 ffs.

 

[spartandre217]

That's why you shouldn't judge on stats, but judge performances on the pitch.

Anway I don't think Janssen is good enough yet. Like Huntelaar and Horseface he started slowly (2nd tier), but has been developing very well. Could do with one more season in Holland, before moving elsewhere.

You can absolutely gain insight and do first pass elimination of prospects using advanced stats. (Not shallow things like pass completion, goals, assists etc)

I don't know enough about him or the eredivisie to make any judgements but if we do take him it's definitely a StatDNA signing. Just like Elneny was.

Way I see it, if he's generating a high number of shots with high xG values, is overperforming league-strength adjusted xG by a reasonable amount (5-10%), isn't dispossessed often and plays in a team with a similar system as us then I'm fine with it.
Would also be nice if he wasn't as dependent on service as Giroud and didn't get too many goals from glaring defensive errors.

The way "stats" has been used in the past was primarily in a very shallow fashion with little to no focus on context, machine learning or big data. That has since changed and there's a lot more to be gained from it than ever before but it is far far from the be all end all and hopefully will never be.

 

[Dynamic]

I wouldnt rely on an amount of goals scored in the eredivise.
The last top goal scorer was Depay.

 

[spartandre217]

I wouldnt rely on an amount of goals scored in the eredivise.
The last top goal scorer was Depay.

Depay was taking a boatload of shots to get those goals, took direct FKs and PK's to boot.

He scored 22 in the Dutch league. 7 were FK's and 2 were PK's. So 13 -ish goals were from open play. So there's that.

Goals from open play can be considered maybe 50-75% as difficult as goals scored elsewhere so he scored 6-9 PL goal equivalents from open play. Which is quite in line with his performances this year, admittedly in a grim United team and with limited minutes.

So there's that.

 

[razörist]

Scoring free kicks is bad?

 

[spartandre217]

Scoring free kicks is bad?

Scoring FK's isn't exactly a skill or event that is likely to occur a) frequently, b) repetitive over long periods.

FK conversion in the EPL from 2007- 2013 was something like 5.3%. Finishing for all other shots was something like 10% over that same period. To say nothing of the fact that direct FK's aren't nearly as common as other chances to convert. So converting some absurd # of FK's at 10-20% of attempts is *not* repeatable.

Admittedly though direct FK's are more likely to go in that other shots from a similar location but compared to shots *inside* the box they're simply not likely.

So if you manage in a single season to put away 7, either you've gotten 70 attempts from direct FK's or you've gotten somewhat lucky. If you pull off high conversions over multiple seasons then and only then would you consider a player's FK conversion as a reliable part of their normal goal contribution.

It's the same kind of logic for why people *mostly* disregard PK's when it comes to evaluating a player's goalscoring prowess. One year a team could be awarded 11 PK's like Leicester this year, another you could get 2, like we did this year.

A player that scored 21 in 1 season and 12 the next season in the first and 2nd cases are at a shallow level, completely different players but may have played to *exactly* the same level in both.

 

[razörist]

By your logic you would not be able to judge any young player on anything, because they have only showed it for one season. But I'm all for that, that's why there are scouts like de Visser from Chelsea, who knows nothing about stats but uncovered gem after gem.

I get what you're saying, but to disregard it completely is bullshit.

 

[hydrofluoric acid]

So in order to go to England from Holland league as a striker how much goal would one have to score to get your people approval? 50 goals in season? 100 goals? Surely he would get to our bench with 150 goals in one season?

 

[razörist]

So in order to go to England from Holland league as a striker how much goal would one have to score to get your people approval? 50 goals in season? 100 goals? Surely he would get to our bench with 150 goals in one season?

Hahahahahha, please stop taking the piss.

 

[James Bond]

21? He looks about 35 ffs.

Where's Marcus and Mike to make sense of this?

 

[spartandre217]

By your logic you would not be able to judge any young player on anything, because they have only showed it for one season. But I'm all for that, that's why there are scouts like de Visser from Chelsea, who knows nothing about stats but uncovered gem after gem.

I get what you're saying, but to disregard it completely is bullshit.

You can't accurately judge any player when they've only played a single season, especially if a large % of their goals are PK's or direct free kicks or easy chances. Something literally everyone here says *all the damn time*.


There are literally hundreds of thousands of players and games a year. There is literally no way for any scout for a single club in a single country to cover any appreciable %-age of them.

The best scouts have a good idea of where the best are most likely to be at any given point in time. They didn't get that in a single season or single observation, they did it over years and in many different places. Those scouts have mental models(possibly biased ones) that are pretty much the same as the ones the stats people you so denigrate have simply put into digital form.

Re: the scouts, they *never* get chastised for the players they recommend that don't pan out but are praised to no end for any successful ID of a player. The opposite is true of "stats". A single failure is a reflection of all approaches but successes are considered "luck".

I don't think there will be ever be a replacement for scouting by eye any time soon but like I said "stats" is a very very very useful first pass. Then you send in the scouts.

 

[Vinci]

The level of goalkeepers in the Dutch League is kind of embarrassing, so I wouldn't look too much into his stats. This was from last week and this is not a rarity, such mistakes happen nearly every week in the league:

 

[hydrofluoric acid]

What about the level of the premier league?

Leicester signed some no namers from France league to survive relegation, they destroyed the league. West Ham signed some 28 year old who no one ever heard of before that made the team of the season. Tottenham have fish as their striker scoring for fun last two years. At Arsenal there is model scoring 20 goals each season. This league does nothing in Europe.

 

[Trilly]

You can absolutely gain insight and do first pass elimination of prospects using advanced stats. (Not shallow things like pass completion, goals, assists etc)

I don't know enough about him or the eredivisie to make any judgements but if we do take him it's definitely a StatDNA signing. Just like Elneny was.

Way I see it, if he's generating a high number of shots with high xG values, is overperforming league-strength adjusted xG by a reasonable amount (5-10%), isn't dispossessed often and plays in a team with a similar system as us then I'm fine with it.
Would also be nice if he wasn't as dependent on service as Giroud and didn't get too many goals from glaring defensive errors.

The way "stats" has been used in the past was primarily in a very shallow fashion with little to no focus on context, machine learning or big data. That has since changed and there's a lot more to be gained from it than ever before but it is far far from the be all end all and hopefully will never be.

Do you study/work in sport analytics?

 

[razörist]

The level of goalkeepers in the Dutch League is kind of embarrassing, so I wouldn't look too much into his stats. This was from last week and this is not a rarity, such mistakes happen nearly every week in the league:

He was actually having a great season, big big mistake from him, that most likely cost his club safety from going down

 

[GDeep™]

Dutch players will have to move to a better league, make their name and then look to the Premier League.

Like Van Dijk, went to Scotland and made his name.