Jump to Navigation
LOBP is now in archive mode... read more at leagueofbikepolo.com/goodbye.

Podium: Bench Stats. Feedback Requested

I'm looking for some feedback on the stats I'll include in Podium's Bench Format. For each Stat, I need a Name, a short descriptive Label, and an Abbreviation. I'd be interested in feedback about any of this, and if you have any good reasons to add or drop any statistics.

Here's the list so far:

"Goal Scored"

"Shot On Goal"

Shot Attempt
"Shot Missed Wide"

"Goal Assist"

"Shot Blocked"

"Goalie Save"




Along with those, a few things are automatically calculated:

Total Player Play Time

Player Goal Differential

Player Goals For

Player Goals Against

BT for ball turnover penalties?

PM incremented by .5 for penalty minutes?

Why not +/- instead of delta for goal diff?

I like delta because of my engineering nerd background. I'll use +/- since nobody ever knows what I mean when I use delta

why a stat for all scoring attempts (including shots not ending in a goal or even on goal) but no stat for all passes ending in a scoring attempt (including shots not ending in a goal or even on goal)?

Goals: Wow, that player scores a lot of goals.
Assists: Wow, that player passes really well.
The stat you're describing above: Wow, that player plays with people who can't finish and/or gives passes that are almost but not quite good enough to score on.

I think the information gathered from that stat is really ambiguous.

"goals"...doesn't mean quality goals. the player may score on breakaways and open nets but whiff every one-timer and shank every shot under pressure. so to me the "goals" aren't as important as the "type of goal".

"assists"...passing really well doesn't always equal goals...your club doesn't have a newbie day? "chances created" is a great way to view it...and yeah I think you bring up two important aspects of the stat 1. exposing non-finishing teammates/players and 2. exposing teammates/players with poor decision making skills...those not quite perfect passes shouldn't result in a low percentage shot because the receiver shouldn't pull the trigger just because the ball is within swinging distance. exposing players suffering from impulse control and poor shot selection is just another bonus to the stat!

the stat would be ambiguous only if it was taken by itself...with an "OT" (one-timer) stat and maybe even a "W" (whiff) stat it paints a more accurate picture of who the playmakers and finishers are on the court. eventually when we not only record shots and passes but shot spots (red x denoting where a shot was taken on the court and missed and a black x for where a shot was taken on the court and scored) and percentages come into play it becomes more important to be able to match up players who can score off of one-timers and who can set up one-timers.

could be I can't see this objectively as I played point in basketball and sweeper in soccer so I'm already biased toward recognition for the hustlers/playmakers!

side note: did you notice there were only three players with positive possesion triple doubles (double digits recorded in three positive stats and zero turnovers or at least more steals than turnovers)? koyo, dusty and shitty...but not one of the three had double digit assists (shitty with 5, dusty and koyo with 3). so did they just not pass, make shitty passes or had shitty teammates? considering it was a merit-based draft shitty teammates should be eliminated from the choices...in theory...which leaves koyo, dusty and shitty looking like they lack the skill, impulse control or court vision to make a pass which I think we can agree is not true. so this stat would have been illustrative of their playmaking prowess or lack thereof...

I'm still not seeing the reasoning here. A goal is a goal, ugly or not. No sport has a column for ugly wins and pretty loses though we've all seen them in a number of sports. At the end of the day, ugly or not, a goal is a goal. And it helps your team to score.

I feel the same way about assists. I sort of get your point about how stats should speak to the quality of player, but shouldn't stats also represent the quality of the team? After all, it's a team sport. So if you're passing well and your teammates are shooting well, then it should be reflected in your assists and their goals. Pick solid teammates and all will play out.

To your last point about how those three dudes were illustrated by the stats, well, it could be they didn't pass much, they didn't pass well or their teammates didn't receive passes well. But it could be none of those scenarios.

Stats are in place to tell you what a player did, not always how they did it.

basketball coaches understand the importance of the types of goal a player can score...dunks but no jump shots, 3 pointers but nothing in the paint, cleanup but no power moves to the rim, etc...this makes a difference in who you team up on the court. we play with teams of 3 now but I'm truly hoping it will transition to bench squads of 5 for 3v3 or 7 for 4v4 (having more than that for either just allows full line changes which is just 3v3 with teams squaded up...that's just 3v3?).

exactly...but we have a chance to decide what stats we think are important before innovation or evolution become mired in tradition. we could have stats that do tell us HOW as well as WHAT a player did...

i dont kno man , bball you can earn either 2 or 3 points depending on how u score so maybe thats why they care to such extent.

but also if the rest of the stats are done properly i think it can tell a lot on how the game was done , and it will take us some how to learn how to read those stats and understand what they mean( a lil bit how u did with the dusty/koyo/shitty exemple) so i say whatever we do we should be able to gather a heaps of intel from any decent stats, we just have to find a way to make them easy to apply and relate to.

also the stats will promote a type of play, if you base a lot of stats on individual stuff it might push people to fill in those stats, or those one.

i mean i rather see stats on how many goals a team is able to achieve in one match rather than how many one timer that dude is able to pull of. if its SO AWESOME people will carry the legend anyway

EDIT : i think the quality of the goals can be "statistified" with a well balanced "defensive move" and "assissts" stats. in theory you should see a team is easy to score on if they dont get a lot "DM" marks against a team thats scored a lot with individual plays ,ie not a lot of "assissts" marks. combine this with the scoring precision of each player and you can basically see the whole game matrix style

Stats mean more or carry more weight when you get to see the goal, assist, block or takeaway. By that, I mean when we can watch footage. And that's a different conversation in itself. I don't think we need the stats to be telling us everything, down to the way a player achieved a goal (breakaway, one-timer, assisted an opponent in an own goal).

I'd like to think video will become more prevalent and lend its hand to stats.

From a usability point of view, if you just record Shots and have a checkbox for Shot On, then Shot Off could be a calculated stat. Also, when you credit a player a goal, the software could automatically add a shot and shot on goal to their totals so you don't have to enter it twice.

I do not see the point of keeping track of shots off goal or passes that could be assists.

Also, just do shots blocked. I do not see the need for blocks and goalie saves.

Regardless, you are doing wonders for the sport, Vince. And I'm super thankful for it.

(shots ON) divided by (shots ON + shots OFF) = accuracy percentage. Rad!!

IMO, tracking (passes that could be assists) would be next to impossible without several statisticians who are all in agreement as to what exactly that is. (I mean, really: "you shoulda scored that one, man! Quit fucking up my stat line!")

Also, I personally don't find tracking dabs to be that important, considering they will likely have the most tallies (at least if it's my stat line) and take statistician focus away from more important stats. But, more stats are cool..

I'd say an accuracy percentage should just be shots IN/shots ON. What's the shots off stat telling me that the shots on isn't? Honestly, I don't care how much a player is shooting as long as he/she is scoring. Nor do I care with the missed shots go. I just wanna see production outta teammates.

Dabs are super important. Staying on your bike is one of the best things you can do.

But I agree about tracking missed assists.

EDIT: Didn't see your comment below, Mr. Do. I really dig what you're saying there. I could see what Vince is doing now leading us to that. It's not impossible, but it's not easy! Hell, none of it is, but I like that direction!

a shot off equals a turnover (the majority of the time) and turnovers equal possession and possession equals goals. if someone scores 4 goals but gave the other team the ball 10 times and the other team scored 5 goals off of those turnovers (that aren't counted as turnovers) I'd say that's an aspect of that players on-court ability I'd find relevant when building a team.

This proves my point: Don't count shots off. If a shot off does result in a turnover, then mark it as a turnover. If it does not, then big deal, your team keeps the ball and nothing came of the play.

but it's not a turnover/steal...that's why it's a stat of its own. are you proposing a loose ball stat? I approve if that...I love stats that reward hustle players and not just selfish players.

A missed shot that is NOT on goal and gets possessed after said missed shot by the opposing team is a giveaway in my estimation. Whether you meant to or not, you're giving the other team a ball.

I'm a huge baseball fan and I'm not sure any sport has more nerds working hard to record stats: temperature at gametime splits, scenario splits (e.g. men on X base w/ Y outs), batting averages when the ball is just hit in play and millions more.

As much as I love looking at a lot of those stats, I feel like hardcourt polo should start small, but think big. So in some regards, I really dig the way you're thinking. I just think for now, Vince should keep it simple and we'll expand from there.

(Also, I'm excited to chat with you about this in person when I arrive in Seattle on Tuesday.)

I think we have to Record shots off. Shooting the ball is way different than a turnover. The end result is only sometimes a TO. Basketball players aren't credited with a TO after a shot. Also its gives lots of insight. What it shows is a player who takes too many low percentage shots or isn't getting results when given the chance to produce or vice versa an efficient and smart player.

- Sincerely
Olsen Aviles

Hey Vince first and foremost I want to thank you for putting so much work into improving our sport.

Second as someone who kept stats for an entire weekend I feel chances created is a very important stat. When we saw someone feed an inch perfect pass only for it be screwed up the shooter all we could do was mark a shot attempt for the shooter. Absolutely no recognition went to the person who made it all possible.

The second stat we felt we should have added was "dribbles". Beating someone one on one happens less often then you think in polo. When someone gets juked and left in the dust by 2 wheels then I feel we should give the guy or girl some credit. Even if it doesn't end in a goal or even a shot Its a great way to objectivity gauge ball control.

- Sincerely
Olsen Aviles

wow with all these stats this sure is gonna make me look great on paper rather than hearing "i didn't see you score any goals Dad"
Thanks everyone

"So this is how it ends"MACHINE

I think 'shot blocked' and 'goalie saved' could be combined. If the last man/girl prevents a goal then count it as a save. This could then be used if they force the attacking to make an error other than a wide shot - thus preventing a goal. Otherwise they will only receive defensive credit if the attacking player shoots.

I agree that it would be great to automatically add a 'shot attempt' and/or 'shot on target' to a player when you record a goal for them.

I like the arguments for simplicity, especially in the early days of this.

I'm leaning towards dropping Shot Attempt, and just having "Shot" - goals automatically count a shot. That would mean that Pass wouldn't be there, since Shot Attempt is gone.

Also having just "Block" instead of Save/Block

Yes, simplicity is king. The sport is about 15-years-old, so it's at a very awkward, acne-ridden, nervous-to-ask-anyone-to-homecoming phase.

And you're helping us make sure we get more than just a hug at the door at the end of the night, Vince.

but, but... shot accuracy percentages!!!

Mr. Do wrote:

but, but... shot accuracy percentages!!!

if he just records shots, you can see how many goals they make, and figure the shot accuracy pretty well. i think. i don't know what i'm talking about most of the time. i just like typing.

I'm not certain about the area around Steal / Turnover. They'd be interesting to have, but it's a little subjective, since someone has to determine the cause - was it a loss by the offensive player, or was it caused by the defense.

I like kev's idea of just counting BT Ball Turnover Penalties. That takes out the subjective aspect..

Penalties are another area. I'd be nice to track the time that a player spends in the penalty box. I could track it like a shift, and you'd be able to visually see if that resulted in a goal.

Loving the way stats are making their way in the game, i was totally spazing out when i discovered the time line with individual name and dated goals.

about shots: i d like to have the total of the shots taken by one + the number of shots that ended up in nets = pecentage of precision for one player, FPS style. add headshots for fun

about block: agree with the general idea that goalie save is unecessary, block are enough. but also , i see a lot of shots get "prevented" (someone putting his mallet in the way, hooking ) before the ball get even smacked i mean.so maybe a more general "successful defensive move" stat , and maybe a way of seeing clearly when its followed by a goal from the defensive team

also a REBOUND stat would be nice, as i think it influence a lot the game if a team have good vision of the rebound game i think we should promote it with some stats.

stuff like "shot missed wide " ," real nice pass with no conclusion " etc seems unecessary to me, i dont think we should bother explaining with stats that someone is screwing up in your team. the fact that you dont get the other stats will be self ex. i think

ASSISTS : are really cool if you count everything(and maybe include in it some of the not so necessary of the above) like passing, chance created , screen,flashbangs i mean : any type of calculated personal action that your team was able to take advantage of and leaded to a goal.
again only succesful action should be taken in account i think , stats are sort of a pro stuff, and as they say : while noobs practice for succes, pros practice at not failling

as olsen (suicide king) noted - two things we took away from bench minor 4 stats that we did not record

- chances created (highlights players ability to create scoring opportunities)
- dribbles (highlights ball handling skills of player)

and dont forget to automatically calculate shooting percentage! (using total shot attempts - not just shots on goal)

I play for hugs

Will all of these stats be recorded on a timeline? I want to be able to calculate Corsi ratings (http://www.theglobeandmail.com/sports/hockey/corsi-rating/article1210138/). Essentially this is just a version of +/- with a larger sample size, although it can be skewed for players that take a lot of dumb shots.

To me, +/- and corsi ratings are more useful than more granular stats like giveaways, takeaways, blocks, etc. Notice that Global United has few blocks across the board. It wasn't because they were bad goalies, it was because they were too busy scoring goals.

I'd sooner avoid subjective stuff (dekes, assists, scoring chances). Keep it simple.

never heard of that name, but yes, it already calculates that. you can see an example here:


One thing that Just came to mind was possession percentage. But that might require podium to manage the time clock as well which would be something that could hinder a tournament if something goes wrong.

I'm pretty excited about this. If people easily can record stats at all the majors for an entire season then we can maybe record them somewhere and see a huge amount of data.

Bench tournaments would be the only realistic thing for now given the sheer amount of games played in a 3 court 3v3 tournament.

- Sincerely
Olsen Aviles

I think it's easy to have high hopes and want to track every stat possible, but that is extremely difficult. The reason we track what we do, in mpls, is to get these percentages that are neat to analyze.

shots/(shots+attempted)=shot accuracy %
goals/(shots+attempted)=goal %
goals/(shots)=another scoring % stat
saves/(saves+goals allowed)=save %

Regardless of what gets tracked via Podium, I'm very excited that this is happening. Thank you, Vince!