"No vote" counting as "Dislike" now ?

[FR_Player-One]

Something I've been wondering about recently.
I was looking at the stats of my latest races, and noticed something weird.

Before, when we had for example 10 plays, 9 likes, 0 dislike, basically 9 likes and 1 no vote, we would have a 100% rate.
Logic right ? Eveyone voted like, the ratio between like/dislike is 100%, except one player who gave a neutral vote, or refused to vote positively or negatively.

But now with the same exact number of votes we get a 95% rate.
On another race, not played a lot, 3 times, I got 2 likes and one No vote, yet the rate is 83%.
And it's like that for all my races/TDM.

Is it just me? You can check your own creations.

It seems that a "No Vote" is counting as half a "Dislike", which can get discouraging even more.
A "No Vote" is supposed to be.. no vote, not a "half ok half not ok" vote. It is not a vote.

Thoughts ?

Edited April 6, 2014 by FR_Player-One

[LambentBunny]

Done that for a while now it sort of counts it towards the overall rating but not really a dislike.

[budcat]

My rating is all over the place. I have 79 plays on one race, 52 thumbs up, 6 thumbs down. 52 out of 58 ratings is an 89.6% or 90% rounded. 52/79 on the other hand is a 65.8% or 66%. Rockstar says it has a 77% rating.

[Broughy1322]

It seems that a "No Vote" is counting as half a "Dislike", which can get discouraging even more.

A "No Vote" is supposed to be.. no vote, not a "half ok half not ok" vote. It is not a vote.

 

It also counts as half a like too, not just half a dislike, which seems sensible. A race with 9 likes out of 10 plays shouldn't have a 100% rating because 100% of people didn't like it. One person thought it was okay, average, not too bad, not good but not bad. That should be reflected in the rating. It's a more fair representation of how the race is viewed.

[FR_Player-One]

"No vote" means no vote, period. It's the absence of a vote, it's not a "medium" vote, or a "ok vote".

I don't choose to put no vote on someone's creation because I don't "like" or "dislike" it, but because I don't want to give a vote to it and not influence the final rating. Lot of people will do races or TDM and don't think anything about it.

 

As it is now it shouldn't even be called "no vote", because it is a vote. And it's false.

I'm not even sure if you actually don't vote anything if it counts as a "no vote" too.

I said it counts mostly as a dislike because it only influence the rating negatively, by 50% less of a vote, but not positively.

2 people not voting give the same ratio as a dislike, not as a like, in the final rating.
The rating should be counted by the votes of the people who actually voted for or against the track/TDM, not with those who refused to vote.

 

And I'm not even specially talking about my own creations, but others people creations mainly.
Again I don't choose to "not vote" to influence the rating, positively or negatively.

Edited April 6, 2014 by FR_Player-One

[Cuz05]

Agree with Broughy there, if a race is no voted then it can't be 100%. 100% mean everyone likes it.

 

Something I think has changed is that it's now 1 player - 1 vote. I have an older race that is 23 plays, 40 likes. I have a new one that was 8 for 8. Had 9 people out on it last night and they all liked, now reads 13 for 13. Obviously a few of them had raced it before so it looks like their votes were not counted. I have other new races showing a similar pattern.

 

Makes sense as a system overall now, I think.

[GTAMYBABIES]

One word answer: BETA

[Fachuro]

"No vote" means no vote, period. It's the absence of a vote, it's not a "medium" vote, or a "ok vote".

 

I don't choose to put no vote on someone's creation because I don't "like" or "dislike" it, but because I don't want to give a vote to it and not influence the final rating. Lot of people will do races or TDM and don't think anything about it.

 

As it is now it shouldn't even be called "no vote", because it is a vote. And it's false.

I'm not even sure if you actually don't vote anything if it counts as a "no vote" too.

 

I said it counts mostly as a dislike because it only influence the rating negatively, by 50% less of a vote, but not positively.

2 people not voting give the same ratio as a dislike, not as a like, in the final rating.

The rating should be counted by the votes of the people who actually voted for or against the track/TDM, not with those who refused to vote.

 

And I'm not even specially talking about my own creations, but others people creations mainly.

Again I don't choose to "not vote" to influence the rating, positively or negatively.

 

You're looking at this the wrong way, it counts 50% towards a vote, you're not entitled to a like from all players who play your track. The default status as they join your race is the no vote, that counts towards a 50% rating, any dislike is then actually a dislike, and a like is then actually a like. That being said, dislikes aren't really dislikes though as a bunch of people are just being dicks. But can't really do anything about that...

 

What I do hope though, is that the system allow for people to change their vote, since it's 1 like per player, the same should count for dislikes and no votes, 1 vote in general per player. So if I like a track I previously no voted or disliked I can show that I'm pleased with the changes the creator made for example, and one of the no votes will disappear, and be changed for a like.

[KimberJane]

Yep well it's obvious that if you had 10 racers and 9 liked it and the other person pressed no vote then it's 90% of people like it. Couldn't be 100%.

That said it's misleading as you would automatically assume the negative percentage in this case 10% would seem that 10% of people 'dislike' the race.

Rather annoying because it makes people think your race is substandard. How it should be displayed is a full breakdown of percentage on the title screen - likes %/ dislikes %/ no votes % (L/D/NV).

Edited April 7, 2014 by KimberJane

[Fachuro]

Yep well it's obvious that if you had 10 racers and 9 liked it and the other person pressed no vote then it's 90% of people like it. Couldn't be 100%.

That said it's misleading as you would automatically assume the negative percentage in this case 10% would seem that 10% of people 'dislike' the race.

Rather annoying because it makes people think your race is substandard. How it should be displayed is a full breakdown of percentage on the title screen - likes %/ dislikes %/ no votes % (L/D/NV).

It doesn't show 90% if it's a no vote, it shows 95%. No vote counts as a 50% vote.

[LuapYllier]

Where are you guys getting this half vote stuff? Is there a source?

 

What I know for a fact is that every like and every dislike is recorded...but every play is not. If your job has been played within a playlist only the playlist receives the "play" but the individual jobs receive the votes.

 

My understanding is that a novote does not count as anything at all.

 

Example. I have a deathmatch that is rated 80% . It has 10 likes, 1dislike, 12 plays.

10 likes divided by 0.80 is 12.5

If you round to 13 this is the total plays true number. 10/13 = .769, or .80 rounded which is 80%

Likes plus dislikes is 11 which leaves 2 no votes.

 

I know this to be correct as this job has been played exactly once in a playlist. This accounts for the one missing play. I have posted this info before and at the time I had several races with low enough stats to be positive of the number of times they had been played.

 

If any of you have a job that you know for a fact that has never been played within a playlist then that is the one which would truely test this stat.

[Cuz05]

I've picked up plays from PLs Luap, almost all my races are played in them.... A lot of them have only been played once or twice as individual races. Am I missing something?

[Broughy1322]

Where are you guys getting this half vote stuff? Is there a source?

 

What I know for a fact is that every like and every dislike is recorded...but every play is not. If your job has been played within a playlist only the playlist receives the "play" but the individual jobs receive the votes.

 

My understanding is that a novote does not count as anything at all.

 

Example. I have a deathmatch that is rated 80% . It has 10 likes, 1dislike, 12 plays.

10 likes divided by 0.80 is 12.5

If you round to 13 this is the total plays true number. 10/13 = .769, or .80 rounded which is 80%

Likes plus dislikes is 11 which leaves 2 no votes.

 

I know this to be correct as this job has been played exactly once in a playlist. This accounts for the one missing play. I have posted this info before and at the time I had several races with low enough stats to be positive of the number of times they had been played.

 

If any of you have a job that you know for a fact that has never been played within a playlist then that is the one which would truely test this stat.

 

I have no doubt you're correct about plays not showing from playlists and whatnot but that bolded part is wrong. 0.769 is 76.9%, which would be rounded to 77%. Not 80%. Rounding up by 3% would be a bit much. It doesn't work out the other way either mind since counting no votes as half a vote would give around 84%. Although if you'd had 15 plays (3 people in that playlist run?) then 10 likes + 4 no votes would be exactly 80% if no votes counted as half. Either way it's all very strange.

 

I currently have 15 plays with 15 likes on this track: http://rsg.ms/QYwkmZ (can confirm those numbers are correct, no hidden plays). If someone apart from Luap wants to play it (since he already has) and leave a no vote to see what happens feel free. Would suggest only one person doing it though so it doesn't get more confusing (and them mentioning it here so everyone knows).

Edited April 7, 2014 by Broughy1322

[LuapYllier]

hmmm. Yeah your right...brain fart on that one.

 

So now I am lost again.

 

Honestly I'm not sure it is completely accurate at all since half the other sh*t is broken.

If you had PM'ed me with that we could have tested if a vote changed is removed from like and placed in dislike. Interestingly...I played that track in a playlist when I judged it.

Edited April 7, 2014 by LuapYllier

[Broughy1322]

hmmm. Yeah your right...brain fart on that one.

 

So now I am lost again.

 

Honestly I'm not sure it is completely accurate at all since half the other sh*t is broken.

 

If you had PM'ed me with that we could have tested if a vote changed is removed from like and placed in dislike. Interestingly...I played that track in a playlist when I judged it.

 

Oh I see. Well after you'd played it it had 1 like and 1 play, so it seems plays from playlists do count. Plays in general seem to be a bit broken, or at least they were, so it would be no surprise if the ratings are too. It's still on 15 for 15 if you want to test that vote change.

 

 

EDIT: Crunched the numbers on some of my own tracks and the no votes counting as half correlates best out of the possible options (with some being bang on) although some of the higher played ones are a little more inaccurate. It's all a mystery. Would have to do some proper testing between a group of 3-4 people and some dedicated "never played before" tracks to really figure it out.

Edited April 7, 2014 by Broughy1322

[Cuz05]

It may not be registering plays in a uniform way. Last night I had 7 people out on that track I mentioned was 13 for 13, in a list. I believe at least 4 of them hadn't run it before but I only picked up 3 on both numbers. Would have to ask a couple whether they had raced it previously to be 100% sure.

So I guess its actually not making sense as a system after all...

[Cuz05]

Another race of mine, on 2 likes, 3 plays from the 1st test run. Run for the 2nd time last night with 5 people that I know for sure have not played it, all liked, race picked up just 3 likes/plays. So its definitely not registering properly.

Race stats are 5 likes, 6 plays, 92%. 1st run was as an individual race, 2 likes, 3 plays. Accurate return. 2nd run, in a playlist, 2 that'd run it before, 5 who hadn't, all 7 liked. Don't know what you can work out from that apart from 'its a bit fkd up...'

Edited April 8, 2014 by Cuz05

[Cuz05]

5 likes, 1 no vote on 6 plays does work out at 92% if you assign a half value to the no vote.

Like to know where the additional plays went tho...

[Fachuro]

Where are you guys getting this half vote stuff? Is there a source?

 

What I know for a fact is that every like and every dislike is recorded...but every play is not. If your job has been played within a playlist only the playlist receives the "play" but the individual jobs receive the votes.

 

My understanding is that a novote does not count as anything at all.

 

Example. I have a deathmatch that is rated 80% . It has 10 likes, 1dislike, 12 plays.

10 likes divided by 0.80 is 12.5

If you round to 13 this is the total plays true number. 10/13 = .769, or .80 rounded which is 80%

Likes plus dislikes is 11 which leaves 2 no votes.

 

I know this to be correct as this job has been played exactly once in a playlist. This accounts for the one missing play. I have posted this info before and at the time I had several races with low enough stats to be positive of the number of times they had been played.

 

If any of you have a job that you know for a fact that has never been played within a playlist then that is the one which would truely test this stat.

 

That's how it used to work, but after they managed to get plays up and running to update almost instantly like it does now, it started recording only 1 play and 1 vote from each player. The 50% vote is based on the fact that 9 likes and 1 no vote gives 95% rating whilst 9 likes and 1 dislike gives 90% rating. This means if a player who previously played your job plays it, their plays and likes won't be counting on social club, but they still get RP for rating it under the new system. The old system is what you're describing, which is the reason so many jobs have 100 likes and 60 plays and such, because the same people could vote like over and over whilst their plays weren't recorded. And I suspect it wasn't the fact it was a playlist, but that those playlist usually consisted of people who had played the job before, but since the likes was broken you would still be getting likes from them.

 

EDIT: And before you make a ton of other theories, this is how all my jobs have been consistently updating with plays/likes over the last few weeks, since that change was made. I was confused at first, but kept watching the numbers, and it always give me the number of fresh players and fresh votes (like/dislike/no vote). I'm not sure how it handles a player who votes differently then he had before though, but I would guess it goes for the majority vote (so if you voted dislike for the same track 5 times and like 4 times, it will show a dislike from you, if you vote like 2 more times it will lose a dislike and gain a like, just my guess)

Edited April 8, 2014 by Fachuro

[Fachuro]

 

Where are you guys getting this half vote stuff? Is there a source?

 

What I know for a fact is that every like and every dislike is recorded...but every play is not. If your job has been played within a playlist only the playlist receives the "play" but the individual jobs receive the votes.

 

My understanding is that a novote does not count as anything at all.

 

Example. I have a deathmatch that is rated 80% . It has 10 likes, 1dislike, 12 plays.

10 likes divided by 0.80 is 12.5

If you round to 13 this is the total plays true number. 10/13 = .769, or .80 rounded which is 80%

Likes plus dislikes is 11 which leaves 2 no votes.

 

I know this to be correct as this job has been played exactly once in a playlist. This accounts for the one missing play. I have posted this info before and at the time I had several races with low enough stats to be positive of the number of times they had been played.

 

If any of you have a job that you know for a fact that has never been played within a playlist then that is the one which would truely test this stat.

 

I have no doubt you're correct about plays not showing from playlists and whatnot but that bolded part is wrong. 0.769 is 76.9%, which would be rounded to 77%. Not 80%. Rounding up by 3% would be a bit much. It doesn't work out the other way either mind since counting no votes as half a vote would give around 84%. Although if you'd had 15 plays (3 people in that playlist run?) then 10 likes + 4 no votes would be exactly 80% if no votes counted as half. Either way it's all very strange.

 

I currently have 15 plays with 15 likes on this track: http://rsg.ms/QYwkmZ (can confirm those numbers are correct, no hidden plays). If someone apart from Luap wants to play it (since he already has) and leave a no vote to see what happens feel free. Would suggest only one person doing it though so it doesn't get more confusing (and them mentioning it here so everyone knows).

 

 

You're doing your math wrong as well. if there's 12 plays, 10 like, 1 dislike and 1 no vote:

 

Each individual vote should then account for 100/12 of the rating, which is roughly 8.33%.

So 10 likes should account for 83%, and the no vote should then add another 4.17% to the rating which should give us a rating of 87.17%. The other 4.17% should count towards the dislikes which then should amount to 12.50% of the votes. Adding this up gives us a total of 99.67% where the remaining 0.33% are lost because we are forced to round our numbers at the 2nd decimal for simplicity. Which means if Luap's match have a rating in close range to 87.17% this proves our hypothesis. If it doesn't, the thesis aren't disproved still as the ratings have been messed up by the glitches before, and to be able to completely disprove this thesis we need to test the theory with at least 3 jobs created after the update.

 

As for Luap, I don't know where the f*ck you got that math from, but I hope you had some coffee before getting in your car after making that calculation The total number of plays is a constant, not a variable, and won't change by any type of math. If it says 12 plays it means 12 plays.

[Fachuro]

Sorry for all the posting, but just figured I could use some personal jobs as examples here:

 

Recorded stats as of this moment:

 

http://rsg.ms/1qnjd9e - The Gynecologist

 

8 like

1 no vote

0 dislike

9 total

90% rating

 

http://rsg.ms/OHXwW2 - Rear Up

 

13 like

1 no vote

3 dislike

17 total

75% rating

 

http://rsg.ms/1h2yjfn - The Trucker

 

74 like

2 no vote

4 dislike

80 total

91% rating

 

 

I picked tracks with an exponentially rising amount of plays to strengthen my induction, here's the maths:

 

9 total gives us 11.11% weight per vote

17 total gives us 5.88% weight per vote

80 total gives us 1.25% weight per vote

 

So this should give:

 

88.88% rating

79.38% rating

93.75% rating

 

Conclusion: Whilst the resulting numbers appear within a fairly close range to the originals the fluctuations are varying on both sides of the original numbers, which indicates our hypothesis needs to be further developed. However, the close proximity to the original numbers in all 3 cases indicate we're getting close. (Please note that the overshoot on the 2nd and 3rd job happened even when counting ONLY the likes) My next development of the theory will be harder to prove though, and follows my previous theory on the ability to change your vote from like to dislike and vice versa. I would suspect at this point, that a players vote is accumulated by his collective set of votes on the track in some way, so for example if you have 2 dislikes and 1 like on the same track, your dislike would only count for 0.67% of your total vote. This is just a hypothesis though. Another thought is that it may be caused by those not rating(people not voting for either of the 3 options and just quitting), I have no idea how those players are accounted for though...

 

EDIT: The accumulative theory would make sense as I've had crew members who occasionally no vote or dislike tracks they usually like, and the fact that the actual number is lower then indicating by the maths then indicates no vote and dislike options accounting for more then their "fair" part of the equation, thus suspecting a player who voted like 9 times and dislike 1 time would only count as 90% of a like and 10% of a dislike is plausible. This can only be tested in a controlled environment though. We might need to make a test track specifically for this purpose where we count the effect of our own votes.

Edited April 8, 2014 by Fachuro

[LuapYllier]

I currently have a brand new race that has 2 plays...2 likes...83% rating.

 

Go figure.

[handcuff_charlie]

Something I've been wondering about recently.

I was looking at the stats of my latest races, and noticed something weird.

 

Before, when we had for example 10 plays, 9 likes, 0 dislike, basically 9 likes and 1 no vote, we would have a 100% rate.

Logic right ? Eveyone voted like, the ratio between like/dislike is 100%, except one player who gave a neutral vote, or refused to vote positively or negatively.

 

But now with the same exact number of votes we get a 95% rate.

On another race, not played a lot, 3 times, I got 2 likes and one No vote, yet the rate is 83%.

And it's like that for all my races/TDM.

 

Is it just me? You can check your own creations.

 

It seems that a "No Vote" is counting as half a "Dislike", which can get discouraging even more.

A "No Vote" is supposed to be.. no vote, not a "half ok half not ok" vote. It is not a vote.

 

Thoughts ?

 

 

People need to just forget about the likes/dislikes of their creations. The game is bugged, the numbers often fluctuate wildly then later get corrected or rolled back.....it really has no bearing on anything.

 

It has been stated with certainty that likes are not an automatic trigger for "verified" consideration. Knowing this, why some of you still obsess over likes/dislikes is beyond me.

 

Just play the game, play your creations, share them, and enjoy yourself, and quit watching the pot boil.

[Fachuro]

 

Something I've been wondering about recently.

I was looking at the stats of my latest races, and noticed something weird.

 

Before, when we had for example 10 plays, 9 likes, 0 dislike, basically 9 likes and 1 no vote, we would have a 100% rate.

Logic right ? Eveyone voted like, the ratio between like/dislike is 100%, except one player who gave a neutral vote, or refused to vote positively or negatively.

 

But now with the same exact number of votes we get a 95% rate.

On another race, not played a lot, 3 times, I got 2 likes and one No vote, yet the rate is 83%.

And it's like that for all my races/TDM.

 

Is it just me? You can check your own creations.

 

It seems that a "No Vote" is counting as half a "Dislike", which can get discouraging even more.

A "No Vote" is supposed to be.. no vote, not a "half ok half not ok" vote. It is not a vote.

 

Thoughts ?

 

 

People need to just forget about the likes/dislikes of their creations. The game is bugged, the numbers often fluctuate wildly then later get corrected or rolled back.....it really has no bearing on anything.

 

It has been stated with certainty that likes are not an automatic trigger for "verified" consideration. Knowing this, why some of you still obsess over likes/dislikes is beyond me.

 

Just play the game, play your creations, share them, and enjoy yourself, and quit watching the pot boil.

 

You're talking to the wrong guys about obsessing over likes/dislikes I think, I don't think Luap cares a lot about it, and I know myself that whilst I admittedly care a little, it's always nice to see people like your creations, I'm mainly about creating new and original stuff. Whatever rating it gets, it gets... What this is about however is understanding how it works, knowledge. If I see something I don't understand, I have to make sense of it and figure it out, that's just how I work as a person. And I also expect the rating system to be a powerful tool in the creation process (seeing what works and what don't once we start understanding how it works, and discover how to properly use it as feedback on our creations.

[CJXJ220]

http://rsg.ms/1gjZmFg

 

5 plays

1 like

0 dislikes

 

60% rating

 

 

I think there is a perfectly reasonable explanation for this:

 

[LuapYllier]

Exactly. Hell I couldn't care less what my votes are lol...I just like to know how things work.

[FR_Player-One]

 

 

People need to just forget about the likes/dislikes of their creations. The game is bugged, the numbers often fluctuate wildly then later get corrected or rolled back.....it really has no bearing on anything.

 

It has been stated with certainty that likes are not an automatic trigger for "verified" consideration. Knowing this, why some of you still obsess over likes/dislikes is beyond me.

 

Just play the game, play your creations, share them, and enjoy yourself, and quit watching the pot boil.

 

Of course I care about the feedbacks I can get about my own creations, but well it's not like I really care about the number of likes/dislikes, most of my creations are rarely played to be honest (because I tend to not share them or host them a lot).

I only have a couple of races that are over 100 plays (one of them that I only hosted twice), so I'm far from running after that.

 

It seems that the new "percentage system" has been active for a couple of weeks and I only noticed now, so... yeah.

 

But my point was that for me "no vote" IS no vote, and before this change, it was.

Never meant to be a 50/50 vote, so this is misleading.

 

This influences the rating of a job without the consent of the people who actually don't want to vote, therefore not influence it...

[budcat]

For someone who does a lot of math to be left with an unsolvable but simple seeming solution is frustrating. I want to know what's at play here. Screw f*ckers who don't like my races, lol.

[Broughy1322]

 

I have no doubt you're correct about plays not showing from playlists and whatnot but that bolded part is wrong. 0.769 is 76.9%, which would be rounded to 77%. Not 80%. Rounding up by 3% would be a bit much. It doesn't work out the other way either mind since counting no votes as half a vote would give around 84%. Although if you'd had 15 plays (3 people in that playlist run?) then 10 likes + 4 no votes would be exactly 80% if no votes counted as half. Either way it's all very strange.

 

I currently have 15 plays with 15 likes on this track: http://rsg.ms/QYwkmZ (can confirm those numbers are correct, no hidden plays). If someone apart from Luap wants to play it (since he already has) and leave a no vote to see what happens feel free. Would suggest only one person doing it though so it doesn't get more confusing (and them mentioning it here so everyone knows).

 

You're doing your math wrong as well. if there's 12 plays, 10 like, 1 dislike and 1 no vote:

 

*snip*

 

Which bit exactly? My maths was based on the assumption that Luap was correct with his track having 13 plays rather than 12. And the only part where I present any new maths regarding that was in my quote of "around 84%". The rest was just correcting a rounding error. Knowing that 10.5/12 gives an 87.5% rating isn't exactly rocket science

http://rsg.ms/1gjZmFg

 

5 plays

1 like

0 dislikes

 

60% rating

 

Makes complete sense if no votes count as half: 1 likes + 4 no votes = 1 like + 2 likes + 2 dislikes = 3 likes out of 5 plays. 3/5 = 60%

[budcat]

 

http://rsg.ms/1gjZmFg

 

5 plays

1 like

0 dislikes

 

60% rating

 

Makes complete sense if no votes count as half: 1 likes + 4 no votes = 1 like + 2 likes + 2 dislikes = 3 likes out of 5 plays. 3/5 = 60%

 

 

I have 79 plays on one race, 52 thumbs up, 6 thumbs down. 52 out of 58 ratings is an 89.6% or 90% rounded. 52/79 on the other hand is a 65.8% or 66%. Rockstar says it has a 77% rating.

Still a little off for me if that's really how the formula worked. 79 plays with 52 up and 6 down is 21 no votes. If you add 10.5 likes and 10.5 dislikes to the numbers you get 62.5 likes and 16.5 dislikes out of 79 total plays. Rounded down 62/79 = 78.4%, rounded up 79.7%, unrounded 79.1%. Rockstar's 77% is still off by a couple percentage points. 61/79 does equal 77.2%, so maybe one of the thumbs up actually isn't counted for whatever reason.

 

Dunno.

Edited April 9, 2014 by budcat