Hi everyone! Thank you so much for the incredible support you've shown for this video. My channel has grown almost 100x as large because of you guys and I am so so so grateful for you all. I'm kicking off Bonby Season 2 with a bang tonight in a new video responding to all of your comments & critiques. Check it out here: czcams.com/video/i-eu4dvh_DE/video.html
I think if will Neff didn't spend most of the final match throwing left-handed, not throwing at all, lobbing the ball up to give people easy catches, etc he would have been 1st team.
I primarily watched the tournament as a PointCrow fan, and I was sure his dodges and that legendary comeback play would have put him pretty high up. It's fascinating to see how much your perspective can get skewed.
Definitely - when I first started this project I thought for sure that Michael Reeves would come out on top. I think my perspective was skewed by the game he played against Twitch. The final results are really not about the specific highlights that get clipped / shared / memorized. It’s a marathon, not a sprint.
Realizing how good Ludwig actually was a cold splash of water to the face. Stung at first, but refreshing. I appreciate the immense work you put into this.
Great video, slight correction: at 5:39, the example doesn’t mean you are 1% better than the other player. Technically you are 1% more accurate, but when compared to another player, you would be 6% better than them, as your accuracy is 6% better when compared to theirs (16.7 * 1.06 = 17.7), with yours being 1% higher raw score. An important distinction, as a player with 33% accuracy would be 200% better than an average player, having a shooting average 16% higher.
I wonder how the stats would swing if we added parameters for dodging a targeted throw as well as time spent on court (vs in the out box) :o I saw another thread about how people thought Michael and PointCrow would've been higher, and I wonder if this would've reflected that because a lot of their clutch plays weren't strikes but dodges
Great point, definitely agree - I’ve heard a few others say this too. Here’s the problem: what counts as a dodge? If the ball comes within 5ft of you? 3ft? 1ft? If you have to physically move to avoid being hit? What if someone just makes a really bad throw, does that count as a dodge for the targeted player? Dodge statistics just open up a whole new can of worms as far as “creative interpretation” goes. I tried to incorporate this at the aggregate level, e.g. Sapnap had the least outs per game, so he gets an award for that. But you’d really have to get into the weeds to decide how many official “dodges” each player had and make some tough calls in a lot of places. Targeting statistics are probably the best alternative bogey in my opinion. Nonetheless, I appreciate the thoughtful comment! Thanks for checking out the video, glad to see some others putting some serious thought into this.
One possible way to get most ‘dodges’ is to weigh each time they were targeted but not hit by how accurate the thrower is (ie. more points if the thrower had a high accuracy score)
@@b0nbyy You don't need to define a dodge as anything more than "didn't get hit when targeted" and you're already interpreting who the target was. if the average accuracy is 25%, a player that gets hit 20% of the time they are targeted is probably dodging more than a player getting hit 30% of the time. We don't need to know which ones are "actual" dodges and which ones were just gimmies...it should roughly even out in the aggregate in the same way that we didn't worry too much about which strikes were "actual" strikes and which ones were just gimmies.
@@b0nbyy One method to control for the quality of throws is using the sum of targets taken multiplied by each thrower's accuracy rating. The quality of the throws is averaged out in the accuracy score anyway, so someone who dodges a target from a high quality thower like Slime should be awarded a higher score for that miss beacause the cost of slime missing a shot is higher than another player with a lower average accuracy. By incorperating the average quality of throws you can approximate the average quality of dodges. So a player who's targeted by a playing with 10% accuracy and no hit can be awared 0.1 points, because we think that's a lower quality throw anyway, and a player who's targeted with a 20% accuracy and no hit can be awarded 0.2 points because it should be more likely that ball would hit.
@@b0nbyyI think an interesting statistic would be targets/out, just showing on average how much a player needed to be targeted to get out. This is a good way to give kudos to people constantly dodging and showing who is staying on the court soaking up targets
This was an awesome video! Sucks that your channel is so small atm, definitely underrated - if you keep up such entertaining down to earth content I can see you going super far with it! Please please please keep it up, can’t wait to see more videos like this
I thought the same thing - if I remember correctly, it took Will about 30 seconds to get revenge on Stanz for the headshot, so there's that. Otherwise, I think Will definitely would have ranked higher if he hadn't taken it easy on Team CZcams. He intentionally started throwing in the championship game to give Team CZcams a couple wins.
I love this video. Not sure if you already knew this, but when comparing percentages (like at 5:31) you should phrase it as "1 percentage point better." It would help make the stats easier to understand. At 8:30 it isn't clear if Slime is 10% above the average (16.7*1.1=18.37) or if he is 10 percentage points above the average (26.7%) Thank you for all of your effort in this video. Can't wait to watch your future content.
I love this intersection of data/stats and content creators. I imagine the bulk of the time spent here HAD to be manually entering the 2400 data points while scrubbing the video. If only there was an automated way to grab data from streamer events the analysis could be CRAZY.
I think the only thing missing from the point system is a way to include the dodges you are able to do, like Poincrow was targeted 33.3 times a mach and was out only 4.3 times, wich means he was able to waste 29 shots from the other team (the best at that stat) but he doesn't get points for that
I think assists would be interesting as a half point. A miss that was followed immediately by a point for another player. If you were often the aggressor, it creates more opportunities for other players to get points
As a data scientist and youtube addict, i cannot express how pleased this video makes me, very very cool to see your analysis And mega props for doing your own data collection, can be really gruelling so appreciate the effort you put in there too
Im amazed will neff wasnt in the top bracket of players. I feel like as soon as caroline got smacked in the face will just turned into a new breed of player and dominated the game well and truly
I would've crushed them all, and now that I have uploaded a single video I qualify as a creator. My only talent in life is dodgeball, my life is a sham.
Its interesting to see the raw stats like this, cool stuff! Id be fascinated to see a weighted stat system where how many team mates were left vs how many opponents would effect scores for getting strikes and outs. A round where most of your team is alive at the end is definitely less valuable in terms of personal stats vs a round like Michaels with the 1v4 clutch to bring it back. Kinda in my mind how Valorant weights how and when kills & rounds won effect personal score compared to raw stat & team scores. I feel like dodges also would have been a fun stat to see written out as well. As briefly mentioned with the Michael in the YT VS Twitch match with how many times he was targeted vs how many times he was 'struck' resulting in these clutched rounds
as someone who loves baseball stats i think you could do some fun things here. some stats you could implement would be how much someone was targeted vs actually hit, or what was the order someone got out in and how long they usually lasted. you could also see how well particular players do against each other, for example how well ludwig does when targeted by certain players or teams vs others. this is super cool amazing video
Great video! I love this style of stuff, would love to see more if you're into continuing this kind of video. It does take a lot of time, but I feel like you could be the SummoningSalt of streamer-event-statistical-analysis videos
For all you statistics students who are talking about significance: IT DOESN’T APPLY because the sample is not in fact a sample and is instead the population, and it’s not being extrapolated out into general dodgeball or dodgeball for said person, it is event specific.
Shoutout bobny guy is making fire edits. Intro was a bit long but I feel like the intro is the hardest part to get dialed and I can’t wait for more useless statistical analysis videos!
I am a numbers nerd and absolutely loved this video and if you make anymore like it I will watch tf out of them. Also, charts are sexy so think about that for next time. js
Instead of targeted, you should have rewarded dodge percentage/dodge volume to reward players for consistently dodging hits, just like how catch is being rewarded. Targeted unfairly skews the data to show the worse players as being better because if they get out super early, their targeted number would be really low.
This was a really cool little tidbit of statistical analysis. I always hated stats class and never really got the real world application, but this seems pretty cool.
If they do another creator dodgeball or if you do another video with more data I'd love to see throws separated into: Off-Target, On-Target Dodged, On-Target Blocked/Catched & On-Target Hit. This would give a clearer idea to the quality of a players throws and quality of a players defensive ability.
Hi everyone! Thank you so much for the incredible support you've shown for this video. My channel has grown almost 100x as large because of you guys and I am so so so grateful for you all. I'm kicking off Bonby Season 2 with a bang tonight in a new video responding to all of your comments & critiques. Check it out here: czcams.com/video/i-eu4dvh_DE/video.html
"How I secretly made a statistical analysis to trick my chat"
Lmao, I saw a lot of people were saying that in his chat. Can confirm I have never made a single dollar from CZcams or Ludwig lol
@@b0nbyy >> yet
better get this guy on commentary next year
^^^This comment. Problem is I’d be OP asf on the court, they’d have to have me as a coach or a sideline reporter or something.
@@b0nbyy:) ❤
I love how some players clearly had special abilities, like Lily's summon Michael and Will's rage buff when Caroline got hit in the face.
I think if will Neff didn't spend most of the final match throwing left-handed, not throwing at all, lobbing the ball up to give people easy catches, etc he would have been 1st team.
I primarily watched the tournament as a PointCrow fan, and I was sure his dodges and that legendary comeback play would have put him pretty high up. It's fascinating to see how much your perspective can get skewed.
Definitely - when I first started this project I thought for sure that Michael Reeves would come out on top. I think my perspective was skewed by the game he played against Twitch. The final results are really not about the specific highlights that get clipped / shared / memorized. It’s a marathon, not a sprint.
It's funny because I'm the opposite. as a huge Ludwig fan I'm surprised he's so high. I feel like he was sitting out the majority of his games
@@b0nbyyi mean ludwig did run a half marathon, he doesn’t like to mention it but it was 13 miles uphill
Realizing how good Ludwig actually was a cold splash of water to the face. Stung at first, but refreshing. I appreciate the immense work you put into this.
Great video, slight correction: at 5:39, the example doesn’t mean you are 1% better than the other player. Technically you are 1% more accurate, but when compared to another player, you would be 6% better than them, as your accuracy is 6% better when compared to theirs (16.7 * 1.06 = 17.7), with yours being 1% higher raw score. An important distinction, as a player with 33% accuracy would be 200% better than an average player, having a shooting average 16% higher.
youtube knows exactly what i want to see. great video
Thanks! Algorithm is definitely showing it some love. Glad to hear my target demographic is being reached.
I wonder how the stats would swing if we added parameters for dodging a targeted throw as well as time spent on court (vs in the out box) :o I saw another thread about how people thought Michael and PointCrow would've been higher, and I wonder if this would've reflected that because a lot of their clutch plays weren't strikes but dodges
Great point, definitely agree - I’ve heard a few others say this too.
Here’s the problem: what counts as a dodge? If the ball comes within 5ft of you? 3ft? 1ft? If you have to physically move to avoid being hit? What if someone just makes a really bad throw, does that count as a dodge for the targeted player? Dodge statistics just open up a whole new can of worms as far as “creative interpretation” goes.
I tried to incorporate this at the aggregate level, e.g. Sapnap had the least outs per game, so he gets an award for that. But you’d really have to get into the weeds to decide how many official “dodges” each player had and make some tough calls in a lot of places. Targeting statistics are probably the best alternative bogey in my opinion.
Nonetheless, I appreciate the thoughtful comment! Thanks for checking out the video, glad to see some others putting some serious thought into this.
One possible way to get most ‘dodges’ is to weigh each time they were targeted but not hit by how accurate the thrower is (ie. more points if the thrower had a high accuracy score)
@@b0nbyy You don't need to define a dodge as anything more than "didn't get hit when targeted" and you're already interpreting who the target was.
if the average accuracy is 25%, a player that gets hit 20% of the time they are targeted is probably dodging more than a player getting hit 30% of the time. We don't need to know which ones are "actual" dodges and which ones were just gimmies...it should roughly even out in the aggregate in the same way that we didn't worry too much about which strikes were "actual" strikes and which ones were just gimmies.
@@b0nbyy One method to control for the quality of throws is using the sum of targets taken multiplied by each thrower's accuracy rating. The quality of the throws is averaged out in the accuracy score anyway, so someone who dodges a target from a high quality thower like Slime should be awarded a higher score for that miss beacause the cost of slime missing a shot is higher than another player with a lower average accuracy. By incorperating the average quality of throws you can approximate the average quality of dodges. So a player who's targeted by a playing with 10% accuracy and no hit can be awared 0.1 points, because we think that's a lower quality throw anyway, and a player who's targeted with a 20% accuracy and no hit can be awarded 0.2 points because it should be more likely that ball would hit.
@@b0nbyyI think an interesting statistic would be targets/out, just showing on average how much a player needed to be targeted to get out. This is a good way to give kudos to people constantly dodging and showing who is staying on the court soaking up targets
lugwig played this on stream great vid
Thank you!
wait when? is therte a vod to watch?
Really sick video, I love analysis videos like this and you did an amazing job
Thank you so much! Hope you stick around for the next one.
This was an awesome video! Sucks that your channel is so small atm, definitely underrated - if you keep up such entertaining down to earth content I can see you going super far with it! Please please please keep it up, can’t wait to see more videos like this
Thank you so so so much for the kind words. I won’t let you down. New videos are already in the works…
Another certified Bonby banger
Hereby ratified and confirmed. Thanks brotha
Finally the content I’ve been waiting for.
Here from will neffs stream - great video!!
Thanks so much! Will's community is awesome.
10:12 Sapnap has an unorthodox movement pattern from playing in (and winning) so many MCC dodgebolts!!!
To quote the announcer at the time "he's so bouncy," boy's got hops
Was there a notable difference in will's performance after caroline got hit in her face. It felt like it and will also mentioned it
I thought the same thing - if I remember correctly, it took Will about 30 seconds to get revenge on Stanz for the headshot, so there's that. Otherwise, I think Will definitely would have ranked higher if he hadn't taken it easy on Team CZcams. He intentionally started throwing in the championship game to give Team CZcams a couple wins.
Okay so can we talk about how damn good your editing is? You're about to blow up dude
I love this video. Not sure if you already knew this, but when comparing percentages (like at 5:31) you should phrase it as "1 percentage point better." It would help make the stats easier to understand. At 8:30 it isn't clear if Slime is 10% above the average (16.7*1.1=18.37) or if he is 10 percentage points above the average (26.7%) Thank you for all of your effort in this video. Can't wait to watch your future content.
He explained it. He’s using VORP with presents as percentage points above average
You would use "basis points" to avoid confusion e.g. @ 5:31 100 basis points higher than average
Great Video, thanks for creating this. For someone who enjoys stats a lot, I enjoyed the video
If there's one thing I love, it's some good data
nah bro michael is filipino thats why he aint have shoes on, we have a genetic buff
Lmfao fair enough, I did not account for that in the data. Will be sure to factor that in next time.
I love this intersection of data/stats and content creators. I imagine the bulk of the time spent here HAD to be manually entering the 2400 data points while scrubbing the video. If only there was an automated way to grab data from streamer events the analysis could be CRAZY.
Dude this is amazing! Crazy stats!
For the first time I'm glad the algorithm pussed a smaller creator on to my page....
Love the vid keep up the work♥
I think the only thing missing from the point system is a way to include the dodges you are able to do, like Poincrow was targeted 33.3 times a mach and was out only 4.3 times, wich means he was able to waste 29 shots from the other team (the best at that stat) but he doesn't get points for that
Not knowing anything about Basketball is apparently a huge barrier to entry.
I think assists would be interesting as a half point. A miss that was followed immediately by a point for another player. If you were often the aggressor, it creates more opportunities for other players to get points
amazing dedication! here from Ludwig's live :))
Thanks so much! Hope you stick around to see the next video.
crazy pointless stats compiled via spreadsheets purely for silly goofy and curiosity? heck yeah, i loved this video so much
this is fantastic work, thanks for making this video, defintely deserves one of my rare comments!
As a data scientist and youtube addict, i cannot express how pleased this video makes me, very very cool to see your analysis
And mega props for doing your own data collection, can be really gruelling so appreciate the effort you put in there too
earned my sub 100%. you're nuts man, i love this kind of analysis
If they do another one with the draft format someone is definitely gonna come to you and Moneyball their way to an insane team
Thoroughly enjoyed this - the data entry must have been hell!
Great video man! Keep up the quality :)
Will do. Thanks!
Great video! Glad CZcams didn't shit the bed in delivering new content to me for once
here from ludwig stream, sick video
Thank you so much! Appreciate the love ❤️
I love how you can tell you are a stathead and how well done/serious this is, so good
Im amazed will neff wasnt in the top bracket of players. I feel like as soon as caroline got smacked in the face will just turned into a new breed of player and dominated the game well and truly
Hard work went into making this, good job man 👍👍
Yes indeed, glad it paid off!
such a great vid!
Thank you so much!
I can’t wait for more tournaments so we can start getting a Wins Above Replacement stat
𝓵𝓮𝓽𝓼 𝓳𝓾𝓲𝓬𝓮 𝓲𝓽
❤️
Ludwig gonna brag about this for years
i genuinely gasped when i saw that you only had 800 subscribers then immediately subbed
Lmaooo this is soo over the top! I love it!
Lol yeah. Ludwig said I “went a little crazy” with this project and I definitely agree. Thanks for checking it out!
This definitely deserves at least 69,000 views. What a great video. I love this, and hope it gets to at least 420,000 views.
Outro song is legendary takes me back to Mr Sark days
Solid work on the data collection and analyses
Fire editing my guy
Thanks so much! Just recently started editing on Adobe Premiere, glad to hear it’s paying off!
This man has a VOICE and it needs to he heard MORE this was a fantastic analysis and loved to see your algorithms as a math nerd myself 😂❤
I would've crushed them all, and now that I have uploaded a single video I qualify as a creator. My only talent in life is dodgeball, my life is a sham.
Great video, man! I hope Lud takes your advice about implementing a draft if he does another dodgeball tourney.
Its interesting to see the raw stats like this, cool stuff! Id be fascinated to see a weighted stat system where how many team mates were left vs how many opponents would effect scores for getting strikes and outs. A round where most of your team is alive at the end is definitely less valuable in terms of personal stats vs a round like Michaels with the 1v4 clutch to bring it back. Kinda in my mind how Valorant weights how and when kills & rounds won effect personal score compared to raw stat & team scores.
I feel like dodges also would have been a fun stat to see written out as well. As briefly mentioned with the Michael in the YT VS Twitch match with how many times he was targeted vs how many times he was 'struck' resulting in these clutched rounds
Love the jxmmy high roller influence here
W need more videos like this
as someone who loves baseball stats i think you could do some fun things here. some stats you could implement would be how much someone was targeted vs actually hit, or what was the order someone got out in and how long they usually lasted. you could also see how well particular players do against each other, for example how well ludwig does when targeted by certain players or teams vs others. this is super cool amazing video
Exceedingly well put together video man good work
Great video! I love this style of stuff, would love to see more if you're into continuing this kind of video. It does take a lot of time, but I feel like you could be the SummoningSalt of streamer-event-statistical-analysis videos
We Will Watch Your Career With Great Interest.
Even when I turn to the dark side of the force and become Darth Bonby?
the internet needs this, ty
For all you statistics students who are talking about significance: IT DOESN’T APPLY because the sample is not in fact a sample and is instead the population, and it’s not being extrapolated out into general dodgeball or dodgeball for said person, it is event specific.
Great vid!
Thank you! Great viewer.
Bonby rigged the event
Conspiracy theory nonsense. You’re ruining my chances of getting invited to Creator Basketball this year bro chill
Shoutout bobny guy is making fire edits. Intro was a bit long but I feel like the intro is the hardest part to get dialed and I can’t wait for more useless statistical analysis videos!
Awesome video man.. Hope to see this back next year
Oh boy, that outro music hit me with a wave of nostalgia. Great work on the analysis!
this is why QT always says Ludwig is a jock......
Feel really lucky to have found this on my feed, really enjoyed it and keen to see what you do next!
Jxmyhighroller vibes go hard
You’ve definitely found that formula
Great video
Great viewer.
I never comment, but your extra insight on slight improvements were genuine good ideas. Great vid!
The algorithm gods have blessed me again 🙌🏾
Real nice Video!
Thank you!
This was an epic video, i hope ludwig watches it
Goated thumbnail
Ludwig the GOAT
Them drafting players would give me so much EE4C vibes if anyone watched cod zombies on CZcams during the BO3 era. Untouchable times
This makes me think of when ludwig did that internet history video.
this us so wicked!! definitely subbed
Why can’t I like this more than once?
Another B0nby Banger. Based Carolinte GOAT ranking
Thanks brotha, had to keep it a stack for the people
A great video that I didn't know I wanted
Basketball shoes would have been ideal. They're super lightweight and grippy on those floors. Playing barefoot makes my feet ache thinking about it.
I am a numbers nerd and absolutely loved this video and if you make anymore like it I will watch tf out of them. Also, charts are sexy so think about that for next time. js
Thank god i knew squeex was actually goated
Would love to see more data vids like this!! great work
That was a great video! Glad to see you get more exposure with the current number of views, deserved
great video neffL
Thank you!
Super cool video!!!❤️
The question is, how does this effects LeBron's legacy?
Great Video!
Thanks so much! Glad you enjoyed.
Instead of targeted, you should have rewarded dodge percentage/dodge volume to reward players for consistently dodging hits, just like how catch is being rewarded. Targeted unfairly skews the data to show the worse players as being better because if they get out super early, their targeted number would be really low.
Really interesting video! Thanks 🙏
My pleasure - glad you enjoyed it!
This was a really cool little tidbit of statistical analysis. I always hated stats class and never really got the real world application, but this seems pretty cool.
Banger video
If they do another creator dodgeball or if you do another video with more data I'd love to see throws separated into: Off-Target, On-Target Dodged, On-Target Blocked/Catched & On-Target Hit. This would give a clearer idea to the quality of a players throws and quality of a players defensive ability.