PB Winners Analysis 9th August - 19th November 2019

PB Winners Analysis

Introduction:

You may or may not have seen my previous PB Winners Analysis I posted just before the first international break of the season. This work will have some similarities, providing some statistics and analysis of PB Winners from the start of the season. The data have all been manually recorded by myself, which provides us with all 81 match daysso far this season. This also gives us 236 PB Winnersif we are to exclude the 7 goalkeeper wins. Due to having to collect the data myself, there are a small amount of PB matrix elements that are missing, which are corners won, accurate through ball, big chance created, secondary key pass, last man tackles and all goalkeeper statistics.

If you are wanting the data I have collected then I have no problem in sharing this with you. Just drop me a message via Twitter or Slack. (@FI_KBrown)

Overall Summary Statistics:

Overall Comments:

Age:

Average Age: 25.95 years

Oldest PB Winner: 35 Years – Dante v Amiens

Youngest PB Winner: 18 years – Rodrygo v Galatasary

Nothing really too exciting here. The average being almost 26 years old makes sense when you consider 26.5 years is the mid-point between 18 and 35.

Location:

61.44% of PB Wins are from players playing at home.

This is not surprising. You would expect a home crowd to encourage players to attack. Since attacking elements of the matrix score more points, this aligns with the percentage of PB wins at home. However, it is also not uncommon for a player playing away to win PB if they are playing in a dominant team against a much poorer side. We will also be able to see if home advantage differs from each position.

We can see from the image below that playing at home is more beneficial for midfielders. This may be due to midfielders taking more control of a game at home than they would do away from home, having to remain organised. We can also so that defenders are the position that least rely on being at home. This may be because they are more likely to be busier defensively away from home and can benefit from elements of the matrix such as tackles, interceptions and ball recoveries.

Possession:

Average Possession: 60.53%

Lowest Possession: 24% - Adama v Man City

Highest Possession: 84% - Lenglet v Andorra

We can see from the table below that the average possession for the team of the PB Winners is only very slightly different. Although, from this we can suggest that you would require to have more of the ball to be able to win PB, regardless of whether they are home or away. However, we will also look further into whether having more possession is more beneficial depending on position.

Below, we can see that the average possession increases as we move from forwards, down the pitch to defenders. A suggestion for this may be that defenders required more of the ball to make simple passes to build up a base score. Lenglet’s PB win with 84% possession also supports this. Lenglet recorded a base score of 163, as well as scoring a goal and keeping a clean sheet.

Set Pieces:

37.29% of PB Winners take set pieces.

32.10% of PB Winning Forward take set pieces.

60.49% of PB Winning Midfielders take set pieces.

17.57% of PB Winning Defenders take set pieces.

From the statistics, we can see that a large amount of PB Winning Midfielders take set pieces, and very few defenders do so. Could this be just that there are few set pieces takers that are also defenders, or are they just outscored. My opinion would be that there is just a small number of set piece takers, particularly at the more PB favourable clubs. This could also be said for the Forward category. However, returning to the Midfielder category, it would appear that the set piece taking midfielders outscore their teammates and other competition due to the opportunity to increase their number of crosses per game. This then also increases the chance of a key pass or even assist if they are taking corners, and perhaps even a goal or two if they take free kicks.

PB Scores:

Average PB Winning Score: 232.59

Average PB Winning Score – Forwards: 224.67

Average PB Winning Score – Midfielders: 253.33

Average PB Winning Score – Defenders: 218.57

Highest PB Winning Score: 403 – Kroos v Belarus

Lowest PB Winning Score: 79 – Skov v Wolfsburg

Bronze Match Days:

Average PB Winning Score: 180.83

Average PB Winning Score – Forwards: 167.71

Average PB Winning Score – Midfielders: 208.14

Average PB Winning Score – Defenders: 162.72

Silver Match Days:

Average PB Winning Score: 251.39

Average PB Winning Score – Forwards: 238.77

Average PB Winning Score – Midfielders: 273.62

Average PB Winning Score – Defenders: 241.19

Golden Match Days:

Average PB Winning Score: 247.84

Average PB Winning Score – Forwards: 249.96

Average PB Winning Score – Midfielders: 264.12

Average PB Winning Score – Defenders: 228.72

The statistics above provide some interesting information. It shows that the average PB Winning score is higher on a Silver Match Day than a Golden Match Day. Although this first appears strange, as if there are to be more matches being played then it would be more likely a higher score is posted in each position. However, if we are to consider that Silver Match Days, all of the Euro Qualifiers have been played on this type of game day. Therefore, if the averages of each position are higher for the Euro Qualifiers we can suggest that the Silver Match Day averages are being skewed by the Euro Qualifiers.

Euro Qualifiers:

Average PB Winning Score: 260.32

Average PB Winning Score – Forwards: 248.06

Average PB Winning Score – Midfielders: 296.06

Average PB Winning Score – Defenders: 239.61

Highest PB Winning Score: 403 - Kroos v Belarus

Lowest PB Winning Score: 164 – Bernadeschi v Liechtenstein

The Euro Qualifier statistics now show us that the Midfielders are skewing the data for Silver Match Days with a huge average score. Particularly as Toni Kroos’ gigantic PB score is included within this set of data.

Shots:

Overall:

Average Shots: 3

Most Shots: 13 – Ronaldo v Lithuania

Average Shots - Forwards: 4.51

Average Shots – Midfielders: 2.90

Average Shots – Defender: 1.46

Shots on Target:

Overall: 1.59

Most Shots on Target: 6 – Pukki v Newcastle

Average Shots on Target - Forward: 2.44

Average Shots on Target – Midfielder: 1.53

Average Shots on Target – Defender: 0.72

As expected, the further you move forward through the pitch, the more shots and shots on target. We can also now see how much shots and shots on target contribute to each position’s average PB winning score.

Average Shots Points: 9 (3.87% of Average PB Winning Score)

Average Shots on Target Points: 7.95 (3.42% of Average PB Winning Score)

Joint Average Points: 16.95 (7.29% of Average PB Winning Score)

Forwards:

Average Shots Points: 13.53 (6.02% of Average PB Winning Score)

Average Shots on Target Points: 12.20 (5.43% of Average PB Winning Score)

Joint Average Points: 25.73 (11.45% of Average PB Winning Score)

Midfielders:

Average Shots Points: 8.70 (3.43% of Average PB Winning Score)

Average Shots on Target Points: 7.65 (3.02% of Average PB Winning Score)

Joint Average Points: 16.35 (6.45% of Average PB Winning Score)

Defenders:

Average Shots Points: 4.38 (2% of Average PB Winning Score)

Average Shots on Target Points: 3.60 (1.65% of Average PB Winning Score)

Joint Average Points: 7.98 (3.65% of Average PB Winning Score)

Later on, we will be able to see how both shots and shots on target rank in their contribution to a winning PB score.

Goals:

Overall:

Average Goals: 0.97

Most Goals: 4 – Ronaldo v Lithuania & Gnabry v Tottenham

Total PB Winners Scoring At Least 1: 159 out of 236 (67.37%)

Total PB Winners with Hat-Tricks: 15 (6.36%)

Average Goals – Forwards: 1.47

Average Goals – Midfielders: 0.93

Average Goals – Defenders: 0.47

Game-Winning Goals:

Overall:

107 PB Winners (45.34%) have a game-winning goal.

A game-winning goal only accounts for 15.05% of the average PB winning score. Although, if you stack the points that also come with a game-winning goal (goal scored, shot and shot on target), which totals to 88 points, which accounts for 37.83% of the average PB winning score.

Percentage of PB Winners with a Game-Winning Goal:

Forwards: 62.96%

Midfielders: 44.44%

Defenders: 27.03%

Due to multiple forward scoring goals on match days, it is often that winning PB is separated by scoring a game-winning goal. Therefore, like shots you would expect the PB winning forwards to have a higher percentage of game-winning goals.

Assists:

Overall:

Average Assists: 0.47

Most Assists: 3 – Sabitzer v Union Berlin, De Bruyne v Scotland, Kane v Bulgaria, Werner v Mainz, Chilwell v Montenegro & Moreno v Malta

Stackable Assist Elements:

Big Chance Created – 3 points

Key Pass = 6 points

Pass = 1 point

Cross + Accurate Cross = 3 + 4 = 7 points

Through Ball = 3 points

Accurate Long Ball = 2 points

Therefore, it would be possible for a player to score 30+ points for an assist.

30 points = 12.90% of an Average PB Winning Score.

Forwards:

Average Assists: 0.51

Percentage of Average Forward PB Winning Score: 8.90%

Midfielders:

Average Assists: 0.54

Percentage of Average Midfielders PB Winning Score: 7.89%

Defenders:

Average Assists: 0.36 Percentage of Average Defenders PB Winning Score: 9.15%

The percentages of each position would be higher if you were to also count for the stacked PB elements. I have not done so since every assist can amount to a different number of points.

Passes:

Overall:

Average Passes: 53

Percentage of Average PB Winning Score: 22.77%

Most Passes: 165 – Vogt v Paderborn (100% Pass Accuracy)

Forwards:

Average Passes: 30.56

Percentage of Average Forward PB Winning Score: 13.60%

Most Passes: 67 - El Haddadi v Dudelange

Least Passes: 0 – Caprari v SPAL

Midfielders:

Average Passes: 62.47

Percentage of Average Midfielder PB Winning Score: 24.66%

Most Passes: 147 – Gundogan v Estonia

Least Passes: 22 – Payet v Nice

Defenders:

Average Passes: 67.20

Percentage of Average Defender PB Winning Score: 30.75%

Most Passes: 165 – Vogt v Paderborn

Least Passes: 16 – Brosinski v Paderborn

Passing statistics are often a signal of how dominant a team is during a player’s PB win. We can construct a scatter plot to look at this relationship.

The scatter plot suggests there is a positive correlation with a team’s possession and the number of passes made by the individual player. We could also suggest that data points that are higher on the passes, but require less possession are more key players, and dictate a team’s tempo.

Furthermore, we can carry out a scatter plot for passes and their relationship with winning PB scores.

There also appears to be a positive correlation between passes and winning PB scores. Although, if we are to refer back to each position’s percentage of average winning PB score, passes are clearly more significant for defenders than they are for forwards. This finally suggests that passes are clearly important, but perhaps not the only influential element to a large PB score.

Key Passes:

Overall:

Average Key Passes: 2.67 (16.02 points)

Percentage of Average PB Winning Score: 6.89%

Most Key Passes: 10 – Candreva v Parma

Forwards:

Average Key Passes: 2.54 (15.24 points)

Percentage of Average Forward PB Winning Score: 6.78%

Most Key Passes: Stengs v Estonia

Midfielders:

Average Key Passes: 3.56 (21.36 points)

Percentage of Average Midfielder PB Winning Score: 8.43%

Most Key Passes: 10 – Candreva v Parma

Defenders:

Average Key Passes: 1.89 (11.34 points)

Percentage of Average Defender PB Winning Score: 5.19%

Most Key Passes: 7 - Alexander-Arnold v Tottenham

Key passes alone appear to account for a reasonable amount of a PB winning score, particularly when you consider it is essentially 6 points for a single pass, which can also be stacked with other pass types. Therefore, with Candreva’s 10 key passes equates to 60 points, without looking further into these key passes were also crosses, through balls or just simple short passes. Those 60 points were accountable for almost 20% of a huge PB score of 303. Candreva also made 100 points total in crosses (60 attempted, 40 successful), which accounts for 33% of his PB score, without stacking the pass and possible big chance created points. Therefore, he was able to base more than 50% of his PB score on just key passes and crosses due to the points rewarded for these elements. This leads us to look at whether there is a relationship between key passes and crosses.

The scatter plot above shows a positive correlation between number of crosses and key passes. I would also personally suggest that a player that puts in a lot of crosses is more likely to have a high number of key passes, however, a player that makes a lot of key passes does not necessarily make a lot of crosses. Therefore, if you find a player with a high number of average key passes, have a look to see whether this is because of the number of crosses they put in, or is it for another reason, such as they creative ability around the penalty area.

Crosses:

Attempted & Successful:

Overall:

Average Crosses: 3.87 (11.61 points – 4.99% of Average PB Winning Score)

Average Successful Crosses: 1.23 (4.92 points – 2.12% of Average PB Winning Score)

Combined Points: 16.53 (7.11% of Average PB Winning Score)

Most Crosses: 20 – Candreva v Parma, Alexander Arnold v Tottenham & v Aston Villa

Most Successful Crosses: 10 - Candreva v Parma

Forwards:

Average Crosses: 2.35 (7.05 points – 3.14% of Average Forward PB Winning Score)

Average Successful Crosses: 0.65 (2.6 points – 1.16% of Average Forward PB Winning Score)

Combined Points: 9.65 points (4.30% of Average Forward PB Winning Score)

Midfielders:

Average Crosses: 4.88 (14.64 points – 5.78% of Average Midfield PB Winning Score)

Average Successful Crosses: 1.75 (7 points – 2.76% of Average Midfield PB Winning Score)

Combined Points: 21.64 points (8.54% of Average Midfield PB Winning Score)

Defenders:

Average Crosses: 4.45 (13.35 points – 6.11% of Average Defender PB Winning Score)

Average Successful Crosses: 1.30 (5.2 points – 2.38% of Average Defender PB Winning Score)

Combined Points: 18.55 points (8.49% of Average Defender PB Winning Score)

Set-Piece Takers:

Average Crosses: 7.19 (21.57 points – 9.27% of Average Winning PB Score)

Average Successful Crosses: 2.36 (9.44 points - 4.06% of Average Winning PB Score)

Combined Points: 31.01 (13.33% of Average Winning PB Score)

Non-Set-Piece Takers:

Average Crosses: 1.90 (5.7 points – 2.45% of Average PB Winning Score)

Average Successful Crosses: 0.56 (2.24 points – 0.96% of Average PB Winning Score)

Combined Points: 7.94 points (3.41% of Average PB Winning Score)

Crosses have the potential to really increase a base score of a player. I also personally feel the average number of crosses are skewed by the number of players that have won PB and play throughout the middle of the pitch. Thus, not being in the wider positions to cross the ball. Furthermore, we can see that by taking set-pieces you are far more likely to have a higher base score on average due to having the opportunity to put in a cross uncontested. Therefore, we can further suggest that a set-piece taker that also plays in a wide position will have plenty of opportunities to cross the ball, leading to large base score and the potential to stack other elements of the matrix.

Long Balls:

Overall:

Average Long Balls: 3.25 (6.5 points – 2.79% of Average PB Winning Score)

Most Long Balls: 18 – Martinez v Romania

Forwards:

Average Long Balls: 1.51 (3.02 points – 1.34% of Average Forward PB Winning Score)

Most Long Balls: 11 – Insigne v Saltzburg

Midfielders:

Average Long Balls: 4.20 (8.4 points – 3.32% of Average Midfielder PB Winning Score)

Most Long Balls: 14 – Guilavogui v Hoffenheim

Defenders:

Average Long Balls: 4.12 (8.24 points – 3.77% of Average Defender PB Winning Score)

Most Long Balls: 18 – Martinez v Romania

Like dribbles, long balls are also an element of the PB matrix which are showcased in tweets. Maybe because they are both elements in which were added in the matrix change. Although, we can see that long balls count for very little of an Average PB Winning Score. Even Vogt making a huge amount of long balls (18) in one match, only accounted for 14.4% of his overall PB score (250). The one point I have taken from this was that it was surprising to see midfielders making more long balls on average than defenders as we would often consider long balls to be associated with centre-backs sending a ball in to a channel for a wide player to recieve.

Dribbles:

Attempted & Successful:

Overall:

Average Attempted Dribbles: 2.29 (2.29 points – 0.98% of Average PB Winning Score)

Average Successful Dribbles: 1.46 (2.92 points – 1.26% of Average PB Winning Score)

Combined Points: 5.21 (2.24% of Average PB Winning Score)

Most Attempted Dribbles: 14 – Messi v Slavia Prague

Most Successful Dribbles: 11 – Insigne v Saltzburg

Forwards:

Average Attempted Dribbles: 2.91 (2.91 points – 1.30% of Average Forward PB Score)

Average Successful Dribbles: 1.99 (3.98 points – 1.77% of Average Forward PB Score)

Combined Points: 6.89 (3.07% of Average Forward PB Score)

Midfielders:

Average Attempted Dribbles: 2.49 (2.49 points – 0.98% of Average Midfielder PB Score)

Average Successful Dribbles: 1.57 (3.14 points – 1.24% of Average Midfielder PB Score)

Combined Points: 5.63 (2.22% of Average Midfielder PB Score)

Defenders:

Average Attempted Dribbles: 1.39 (1.39 points – 0.64% of Average Defender PB Score)

Average Successful Dribbles: 0.77 (1.44 points – 0.66% of Average Defender PB Score)

Combined Points: 2.83 (1.29% of Average Defender PB Score)

You will often see a lot of people highlight the number of dribbles attempted or completed by a player during a strong performance. However, from the average point values we can see that dribbles, both attempted and successful contribute in such a minimal way that in my opinion dribbles are not something I would strongly consider when analysis potential PB players.

Interceptions, Blocked Shots & Clearances:

Overall:

Average Interceptions: 1.03 (5.15 points – 2.21% of Average PB Winning Score)

Average Blocked Shots: 0.19 (0.95 points – 0.41% of Average PB Winning Score)

Average Clearances: 0.90 (2.7 points – 1.16% of Average PB Winning Score)

Combined Points: 8.8 points (3.78% of Average PB Winning Score)

Forwards:

Average Interceptions: 0.56 (2.8 points – 1.25% of Average Forward PB Winning Score)

Average Blocked Shots: 0.06 (0.3 points – 0.13% of Average Forward PB Winning Score)

Average Clearances: 0.42 (1.26 points – 0.56% of Average Forward PB Winning Score)

Combined Points: 4.36 points (1.94% of Average Forward PB Winning Score)

Midfielders:

Average Interceptions: 1.10 (5.5 points – 2.17% of Average Midfielder PB Winning Score)

Average Blocked Shots: 0.25 (1.25 points – 0.49% of Average Midfielder PB Winning Score)

Average Clearances: 0.48 (1.44 points - 0.57% of Average Midfielder PB Winning Score)

Combined Points: 8.19 points (3.23% of Average Midfielder PB Winning Score)

Defenders:

Average Interceptions: 1.47 (7.35 points – 3.36% of Average Defender PB Winning Score)

Average Blocked Shots: 0.26 (1.3 points – 0.59% of Average Defender PB Winning Score)

Average Clearances: 1.89 (5.67 points – 2.59% of Average Defender PB Winning Score)

Combined Points: 14.32 points (6.55% of Average Defender PB Winning Score)

These collective defensive minded PB elements do not seem to have much of an impact of PB wins. It is only the defenders that really have any significant impact by these elements. Therefore, I personally wouldn’t worry too much about these elements when looking to research a player.

Tackles & Tackles Won:

Overall:

Average Tackles: 1.66 (4.98 points – 2.14% of Average PB Winning Score)

Average Tackles Won: 1.03 (4.12 points – 1.77% of Average PB Winning Score)

Combined Points: 9.1 (3.91% of Average PB Winning Score)

Most Tackles: 10 – Andre v Angers

Most Tackles Won: 7 – Andre v Angers

Forwards:

Average Tackles: 1.12 (3.36 points – 1.48% of Average Forward PB Winning Score)

Average Tackles Won: 0.68 (2.72 points – 1.21% of Average Forward PB Winning Score)

Combined Points: 6.08 points (2.71% of Average Forward PB Winning Score)

Midfielders:

Average Tackles: 2.06 (6.18 points – 2.44% of Average Midfield PB Winning Score)

Average Tackles Won: 1.23 (4.92 points – 1.94% of Average Midfield PB Winning Score)

Combined Points: 11.1 points (4.38% of Average Midfield PB Winning Score)

Defenders:

Average Tackles: 1.80 (5.4 points – 2.47% of Average Defender PB Winning Score)

Average Tackles Won: 1.19 (4.76 points – 2.18% of Average Defender PB Winning Score)

Combined Points: 10.16 points (4.65% of Average Defender PB Winning Score)

Despite a successful tackle stacking up to a total of 7 points, due to the lack of them in a single game for a player they do not contribute very much towards the larger PB scores. Tackles are also something that are required once the team are not in possession, and PB winning scores are correlated positively with possession. Therefore, this would seem rather counterintuitive to invest in players that make many tackles.

Ball Recoveries:

Overall:

Average Ball Recoveries: 5.74 (17.22 points – 7.40% of Average PB Winning Score)

Most Ball Recoveries: 16 – Rico v Southampton

Forwards:

Average Ball Recoveries: 3.63 (10.89 points – 4.85% of Average Forward PB Winning Score)

Most Ball Recoveries: 9 – Insigne v Fiorentina & Perez v Southampton

Midfielders:

Average Ball Recoveries: 7.05 (21.15 points – 8.35% of Average Midfielder PB Winning Score)

Most Ball Recoveries: 15 – Fred v Astana

Defenders:

Average Ball Recoveries: 6.62 (19.86 points – 9.09% of Average Defender PB Winning Score)

Most Ball Recoveries: 16 – Rico v Southampton

Ball recoveries are a more frequent part of the game for midfielders and defenders, which also account for more of their respective average scores. You can also see that average ball recoveries are almost greater than the total number of points earned for all other defensive elements combined. Therefore, I feel this is an element that is often undervalued in comparison to other elements. When you compare ball recoveries to dribbles, they are far more frequent in a single match and one ball recovery is equal to one successful dribble.

Clean Sheets:

Overall: 71.62% of defenders won PB with a clean sheet bonus (25 points).

Golden Match Day: 64% with clean sheet bonus.

Silver Match Day: 90.63% with clean sheet bonus.

Bronze Match Day: 47.06% with clean sheet bonus.

It is interesting to see that defenders are far more likely to have a clean sheet bonus within their score on a Silver Match Day than any other match day. Although, it may be that due to having internationals and Champions League fixtures on Silver Match Days as the clean sheet bonus makes the difference between winning PB dividends and not so.

Wins:

Overall: 91.10% of PB Winners also had the win bonus points

Forwards: 93.83% includes a win.

Midfielders: 90.12% includes a win.

Defenders: 89.19% includes a win.

Golden Match Days: 93.51% includes a win.

Silver Match Days: 99% includes a win.

Bronze Match Days: 74.58% includes a win.

The winning percentage was always expected to be extremely high. The lower percentage on the Bronze Days also make sense if you think that on some occasions there may only be 1 match and so if that ends up in a draw then nobody playing that match day will have a win bonus.

Additional Analysis:

After looking at a range of PB components individually. I will now create a table for each position of the average points scored from each component in ranking order for each position. Hopefully this will help you to refine your PB research for the future, looking at which elements are key overall and for which position.

Although we have covered a lot of different elements of the PB matrix and looking at how they have had an effect on each position, there is a final factor in which I have not covered and could do with some further research, which is whether a player plays out wide or through the middle. Such as full backs, wide midfielders and wingers would be classes as wide players, with the remaining being classed as central players. If this is something you would like to see in the future then perhaps I can find a way to incorporate that into my analysis.

I have also tried to remain unbiased towards any particular players. At the end of the day, all of my current research is based on historic data and isn’t a prediction on how individual players will perform in the future.

I hope that you have gained some kind of knowledge from this post. I will be looking to post out content that is similar on a more regular basis. If you have any suggestions and/or feedback then that would be greatly appreciated.

If you wish to contact me at any time you can do via Twitter:

@FI_KBrown