Which one of the following do you prefer?
you seem to be ignoring one thing. having "double the odds" excludes the fact that you can't enter as many... so the effective rate is not increased. in other words, having 3% chance to win one GA vs 1% chance to win on 4 GA (of the same game)...
Comment has been collapsed.
You didn't read the last paragraph, did you?
"I can't say if it's better or worse than the previous system yet"
" I'd have to compare how both quantities translate into winning GAs since the odds have changed a lot and there aren't enough finished GAs to do it yet."
Comment has been collapsed.
no need to wait, it's math! Like gravity - not just a good idea - it's law!
Simply think about if you are trying to fill boxes. could you fill more boxes if you get 5 pieces an hour, or enough to fill 5% of the boxes you have each hour? Setting the generation to static where the GA are dynamic means that you will almost always get too few points. *shrug
Comment has been collapsed.
There are no data to prove that. The end game here is winning GAs, not participating in them. We would have to check the odds and how they've changed since the update on a X number of finished GAs to be statistically sure, i.e., with a small enough margin of error so it could be considered legit.
Comment has been collapsed.
The data you are looking for can be inferred here:
(this month is the highest on record, with over 4,500 daily giveaways), and this causes some adverse affects towards user experience on the site. In 2013 and 2014, users received an average of 7,500 points per month. This month users will receive over 45,000 points.
And:
I'm proposing that we set points at a fixed rate of 14,400 per month, which means 480P per day, or 5P distributed every 15 minutes.
Our amount of points have in other words roughly been cut by more than 2/3ds. In order for this to keep giving us equal chances of winning we in other words need to have three times higher chance of winning any individual giveaway. As your calculations show the odds on popular games have barely moved, while the odds on lower interest games have roughly doubled.
I look forward to seeing better data on this over the next month, but so far it isn't looking good.
Comment has been collapsed.
that was the highest
if you look closer it's more around 30k, so it's halfed
Comment has been collapsed.
The curve trends upwards, and the more members the site gets the more giveaways will be created - which in the old system would have meant more points.
Then again it doesn't really matter - if you look at the numbers in the post for what OP calls an "AAA game" you'll see there is barely any change in chance of winning. Everyone chooses to join the giveaways for the more popular games first so the odds just become barely changed for those who actually want to play their wins, while they get better for the card farmers and +1 collectors who doesn't care which games they win.
If it's only halved, as you say, then the odds have to double to be a fair exchange... over the board, not just for less popular giveaways.
Comment has been collapsed.
yes that's what I was thinking while reading this post.
Comment has been collapsed.
Debatable. If you're very selective in what you enter and are constantly sitting on a pile of points, this change affects you positively by increasing you chances. I don't enter giveaways more than I used to, but since entries are reduced, my odds increased.
Comment has been collapsed.
Not only that - even more common example of being not very active. If you are visiting SG only once a day to spend all your points it is also very positive change for you. In old system you would spend 300P every 24h, in new system you spend 480P (60% more) every 24h. So not only you are able to spend actually more points but also for better win chances.
Comment has been collapsed.
Particularly, for the AAA game in question (which happens to be on my WL), you get barely 20-25% higher chances per GA, while being able to enter about 3x fewer (if I followed the point calculation about right). Not a good deal :/
Comment has been collapsed.
if you try a roll simulator for a 100 die , and set cases
First case roll 1 2 or 3 and win 4-100 you lose
Second case roll 1 and win 2-100 you lose
You will come to a result that the 3% are more likely to win cause its a separate roll ,1 roll of 3% is not equal to 3 rolls of 1% statistics dont work like that
I know most people will compare it with 1 GA versus 100 GAs but thats wrong cause in the end till now when you entered 100 GAs with 1% you had that chance... now the next 100 GAs you will enter will have higher chance so you will be most likely to win especially if people start being more selective in what their enter
Comment has been collapsed.
You will come to a result that the 3% are more likely to win cause its a separate roll 1 roll of 3% is not equal to 3 rolls of 1% statistics dont work like that
That's assuming the three 1% rolls are independent, then indeed win chance for three rolls is 1-0.99^3, which is <3% (2.9701%)
But the three rolls are not independent: someone who won won't be participating in later rolls, so the actual win chance becomes: 1-(99/100 x 98/99 x 97/98), which is exactly 3%
Anyhow, even if rolls were independent, the difference is negligible and it's common to round to the sum of those probabilities as long as they're very low and you don't do too many rolls, even when you perfectly know it's not completely accurate.
Comment has been collapsed.
I'm guessing you're just joking but looking at their wins, the only one they've had since march was on a 1 entry giveaway which is probably why they don't consider it a "proper win"
Comment has been collapsed.
Hate it, very slow rate, the best games will still have terrible chances, the only ones with significant difference will be the least wanted games.
Comment has been collapsed.
I wouldn't draw any conclusions from such basic data especially as it can be easilly misinterpret without proper context. Most important context, like you mentioned - would be to compare it with old point regeneration (as well as with point spending, because regeneration alone can be meaningless if you don't look on entering patterns as well). For example if user was able to enter 3-4 times more GAs before the changes then he's entering nowadays then "twice as high chances to win" while sounding nice actually means he will be winning less. On the other hand low activity user who logs in just once a day to spend all his points will win more (before he would spend 300P on lower chances, most of his daily regenerated points would go to waste, now he will spend 480P on better chances and no points will be wasted).
Another thing is that comparing just two GAs of the same game and same timer can give mixed results as well. 1h GA in peak hour vs 1h GA in low attendance hours. 24h GA starting and ending in peak hours vs not (remember that even with 24h GA many users will just scroll through first few pages of GAs before they go out of points, so if GA starts and ends in non-peak hours it may be missed). Day of the week (more users with more time on weekends), in old system were there a flood of points or not (more entries during point floods) etc - these all affect actual results. Thus I believe comparing few random GAs is not really a valid methodology here, even if you compared 10, 20 or 50 GAs it could still be very flawed.
Only thing I really agree with in your OP is "Wait at least a month" ;p This is basically what I am intending to do, just using different methodology. Not looking at GA-GA comparisons but actual numbers for wins which will be much more correct statistically. Draw X random users from community representing different SG levels and activity levels. Compare how much they were winning per X time before changes and how much they are winning after changes. Then you can actually judge what difference it really made ;)
Comment has been collapsed.
This thread was mostly created to stop all apocalyptic conclusion that are based on anything by their personal opnions.
As I said, this is just for fun and I've selected some examples from my data. However you're correct, there is no way to tell how good/bad this update really is without waiting for more data to be able to add all variables, including the different approaches of SGers: Some want a game to play it, others to idle it or just for a +1. Some might not even have one of those reasons. Who knows. There are people who are here every day, every hour and those who barely use the site. Different levels, days, timezones, etc etc etc. In short, there are too many variables to drawn an intelligent conclusion.
Comment has been collapsed.
The amount of wins before and after will always be the same, because there is always exactly one winner per giveaway (assuming one copy). The only way average wins goes up, is if giveaways created goes up.
Comment has been collapsed.
only overall amopunt of wins stays the same, wins distribution is very much subject to change. Let's assume that SG bans group, private, WL and higher levels GAs. While overall number of wins would remain the same (assuming people would start dumping all these games to public lvl0 GAs) wins distribution would change dramatically - high level users and active groups/forums users would start winning like 90% l;ess, while lvl 0-1 userw wopuld rise by small %.
Comment has been collapsed.
This update by design negatively affects bots and +1ers/idlers more than anyone else, so it's strictly positive from my point of view. The boosted chances, whether big or small, are a nice cherry on top.
Also: dog maths 乁( ◔ ౪◔)ㄏ
Comment has been collapsed.
Dogs can't read text hidden behind spoilers, please don't tell LastM I don't actually trust dog stats and that's the reason I didn't use them in my previous reply hi blue weeb \(•‿•)
Comment has been collapsed.
I'm still wondering why people think that this will hurt bots oh so much. Even this thread shows that humans will still be all in on the most appealing giveaways, while the b- and c-selection offer less competition. And who is primarily interested in those GAs with few entries? Bots.
So yes, they'll have fewer points. But they are also those who benefit most from masses of GAs with few entries.
Comment has been collapsed.
Please don't see this as an attack if your opinion differs from mine, I'm just presenting my point of view on the matter ^^
My post said "bots and +1ers/idlers", not just "bots"; basically anyone that entered everything just for the thrill of winning something or for the few cents in steam wallet that came from idling those wins. Those users are definitely the ones negatively affected by the new points regen limitations.
On the flip side, someone that only enters games they'd like to play enjoys a positive change: they now enter the same number of giveaways as before - not getting even close to running out of points - but those giveaways have fewer entries; in other words, giveaways have now a slightly better chance to go to a good home that will play them and love them. That's where my "strictly positive" claim comes from.
Comment has been collapsed.
No worries, I rarely consider a reply an attack. But somehow you either didn't understand my point or you chose to ignore it.
LastM provided some numbers above, showing that the chances to win the most popular games didn't improve much at all. Where the chances dramatically improved, is for less popular games. And who is most interested in GAs with few entries, not caring about which game it is for? Bots and idlers.
Sure, theoretically "selective gamers" might chose to enter such giveaways too. Theoretically. But the numbers clearly show that only a minority does. And anyone who knows the human nature somewhat, can't be surprised by that.
Comment has been collapsed.
I did ignore the point about the "less popular games", my apologies for that.
I just thought that the "beauty" of the fixed regen system is that no matter what, bots and users alike are limited to that upper cap of points, so they have to balance entering a ton of those lower-entries-but-of-limited-appeal low-tier/idle-trash games vs. entering desirable (and usually expensive) games. If bots and idlers decide to spend their points on "crap" they will still win a lot of games for sure, but on the other hand they won't be able to join as many cool giveaways too.
Talking about the numbers provided by LastM you can see a -400ish and -1500ish entries on the two Rise of the Tomb Raider giveaways; to me that change in numbers doesn't look insignificant at all: that's a 15 to 20% of users that decided to spend their points somewhere else, giving higher chances to others. I like Tzaar's example on the matter.
Comment has been collapsed.
Yeah, though Tzaar only covers how many points are available to each. What he does not reflect about at all, is how those points get used.
15-20% less entries does not mean that people chose to enter other games, it far more likely means that they had no points left.
And it seems very doubtful that 15-20% less competition is enough to improve people's chances, considering that they can enter fewer giveaways.
What will ultimately tell us the truth, is this statistic. So far I'm not impressed, but we'll see after some weeks.
Comment has been collapsed.
I think numbers might change in a few weeks, for 2 reasons:
First, the consumers need some time to change and adapt their consumption baskets. It might stay similar at first, but those who just want some cards/+1s for their accounts will probably shift all their entries from AAA to bundled games to keep a steady flow of cards coming their way, since the cost of entering AAA is at a new high, while the cost of entering bundled have been lowered. This shift might take a few months to happen tho.
Second, the biggest group who doesn't discriminate between AAA and bundled games - bots/autojoiners - have their days numbered. According to a mod, the ban hammer is coming. Soon™
So, I believe that in a month the landscape on SG might be really different.
Comment has been collapsed.
I don't think that bots and idlers ever were into AAA games. Form them it always was the smarter choice to go for less prominent GAs.
No idea about the ban hammer. I'll appreciate it if it happens, but that would be an entirely different topic and isn't related to the question if the new point system will make it oh so much more likely to win, for 'selective SG members'.
Comment has been collapsed.
I don't think that bots and idlers ever were into AAA games.
Bots/autojoiners enter everything, they don't discriminate between AAA nor bundled games. If you take a look at some of their groups on steam right now, they are discussing a way to make the bots/autojoiners smarter enough to enter only games with cards in a more randomly way so they don't get caught.
About idlers, a decent amount of them enter even AAA just to idle. In their defense, I don't think they join a GA just to idle, as they seem to care more about the thrill of winning than the game itself. I prefer the term gamblers. I wish I could give you some examples, but that would be against the rules of the site. (it's considered calling someone out) It's not hard to find people like that tho, since some of them are quite famous around here while other live inside groups, only leaving to win public GAs. Just use Kelnage's script and you'll find a bunch of them.
Banning bots/autojoiners might not be 100% related to this question, but no doubt it will change the landscape of this site and in turn complement this update.
As I said, it's not possible to afirm with low margin of error that selective SG members will have a better chance, however, I still think it will be the case for the reasons I pointed out above. the data points to that, but it could be biased, since there aren't enough finished GAs to be sure. Therefore, I could be proved wrong. We'll see.
Comment has been collapsed.
It actually helps bots and +1 ers, since the odds improve the most for the least popular games so they can just enter those giveaways and win more frequently than the people entering giveaways for the more popular games.
Comment has been collapsed.
I won four giveaways in the last 27ish hours, so I really like the new system!
Comment has been collapsed.
Most impeded users are botters.
In the end, I think this is fine too for longer retention of users on the long run, since they won't leave SG from getting burnt out for a too much demanding RNG website.
Comment has been collapsed.
On the contrary botters will have it far easier with the new system than people do.
Considering the new point changes I have to micromanage my points far more than I used to to enter the giveaways I really want.
Bots don't have stamina issues so they can be here all the time working in the background.
Comment has been collapsed.
Most giveaways your average person might enter are throw-away points anyways, this site works well in that even if you give away one game per month, you will most likely always have one game you absolutely want to enter to win.
Comment has been collapsed.
My strategy for now is to enter flash giveaway that i can see or wait for like one hour to end the giveaway and if it is <1500 i enter
Doing this cause on this humble bundle there is two games on my wishlist, maybe some other time there is nothing for me and i can save points, but for now i am carefully spending points on my wishlisted games :D (and i am fine with that)
Thanks for the statistics approach LastM :D
Comment has been collapsed.
i still think dynamic points was better, at least with a minimum-maximum setting so it doesn't either generate too few or too many points.
but overall the update is awesome, limiting entries per user will give more meaning to giveaways entries. hopefully people will start picking games they want to play.
leech-tears are beautiful ;_;
Comment has been collapsed.
I've always entered for the games I really want to play, and I still hate static generation instead of dynamic.
Comment has been collapsed.
I mean, since the chance of winning is way better now the worse the game is the people who don't care about the quality of the game just got a great boost in winchances,
So sure, some people will be more careful with their points and enter only giveaways for games they want to play - but the bots and +1 collectors and card farmers have more incentive now than they did before.
Comment has been collapsed.
Which one of the following do you prefer?
It depends on how frequently you visit the site (or enter giveaways).
Comment has been collapsed.
The good test is tomorrow whith the humble bundle ;) How many time point will be at zero ;)
Comment has been collapsed.
First, I'd need to know the amount of points we were getting every 15 minutes, on average, before the update to compare with the amount we're getting now.
This was mentioned in the first thread discussing the upcoming changes: we were getting 2x to 3x more points in the 3 months before (27k-45k/month vs 14.4k/month now): https://www.steamgifts.com/discussion/kGtHd/point-distribution
That seems to match the 2x-3x better odds on cheap games. Not so much the slight odds improvement on Tomb Raider :/
Actually, the main issue I see with the new system is that point generation doesn't follow GA creation anymore. I think it would have been better to just halve point generation but still keep it related to GA creation
there aren't enough finished GAs to do it yet
True but for AAA games it seems quite clear cut already
Comment has been collapsed.
Also worth noticing that the curve trends upwards, so the disparity in points between what we would have gotten and what we get now will keep growing - making the odds for less popular games ever better - and give people even less chance of winning anything popular.
Comment has been collapsed.
You could probably choose other GAs to prove the exact opposite too..
Just picking specific games and GAs proves nothing..
I'm seeing lots of GAs go unentered by me for games that i want, simply because i don't have the points.
point generation needs to be higher, especially for higher lvl ppl
Comment has been collapsed.
No, he couldn't. Because you won't find GAs that now have higher entry count that before the change. It's impossible. You can't have less points and expect people to spend more of them. At best really desirable AAA GAs that last few weeks will have the same amount of entries. But if you don't agree just go and find single example to counter LastM point.
(No, I won't go and try to find example like this for you. It's your duty to show evidence that will back up your words, not mine).
I'm seeing lots of GAs go unentered by me for games that i want, simply because i don't have the points.
That's whole point of the change. Less entries = higher chance to win. Duh.
Comment has been collapsed.
I actually believe that in this context, proving the exact opposite would be showing "no change", not "having more people in it".
Comment has been collapsed.
if there are
- the same number of users &
- the same number of giveaways
on the site before & after the Change that means:
- the number of winners on the site in a given timeframe is the same.
if a single Person only is able to enter less giveaways than before he/she has to choose more carefully that means:
- the Chance, that the winner will actually enjoy their won game maybe is a Little bit higher.
so.... the new System should be a bit better, shouldn´t it?
Comment has been collapsed.
Well no - because people who collect +1 games and farm trading cards will choose to enter giveaways with better odds - which is everything that isn't popular. The odds have barely changed on the popular games so the people who like popular games will win less than the people who don't care what they win.
Comment has been collapsed.
Most of the whiners are those who win more than they share. I welcome the new system.
Comment has been collapsed.
https://www.steamgifts.com/go/comment/Xa99Css
I put my (probably over-) simplified take on things there.
But, well ... I like to keep things simple.
Comment has been collapsed.
To be honest I don't see what's up with this whole new point system breakdown because it doesn't affect me at all, because I'm cherry picking the games I want to enter.
So if this means fewer pointless entering of giveaways they don't even want to play? Cool. Better odds for those who give a damn.
Comment has been collapsed.
Better odds for people who actually play games
more points for people who care about farming
Comment has been collapsed.
it would had 93 entries before the update!
Comment has been collapsed.
roll back the update cg! it's broken! i need more points! i can't enter everything in my wishlist! it's unfair! this favors casual users! this only favors leeches! this only works for people that don't like to +1! i will leave sg forever! i will cancel my patreon subscription! *rants*
Comment has been collapsed.
Since there is often a very limited number of AAA giveaways, the odds of winning one of these have not improved significantly. The change is most damaging for users which are a) not interested in indie games and b) have otherwise diverse interests, and most helpful for people who only want some specific, unpopular niche game.
Comment has been collapsed.
Got some time to spare, so I decided to compare GAs before and after the update since, I've been seeing a huge difference in number of entries per GA.
Btw, In order to find better results, I'd need more data and same day statistics since entry numbers fluctuate daily. This is mostly just for fun, since there aren't enough GAs for a more comprehensive inquiry. Therefore, don't assume this to be 100% true.
Here are just some examples. My search was a bit bigger, but I'm not gonna list everything. I'm too lazy for that. You can do it yourselves by looking at the archive.
(Format: # entries / % of winning)
The Political Machine 2016 (10P) - lvl 0 / 1h
Before
835 / 0.12%
After
268 / 0.37%
The Political Machine 2016 (10P) - lvl 5 / ~24h* (In normal conditions, I wouldn't use the first GA to compare, since it's not long before the update, but it was the best I could find. Feel free to link me a better one.)
Before
187 / 0.53%
After
91 / 1.1%
The Count Lucanor (10P) - lvl 0 / 1h
Before
441 / 0.23%
After
239 / 0.42%
Let's try a AAA game.
Rise of the Tomb Raider™ (50P) - lvl 0 / 1h
Before
2583 / 0.04%
After
2117 / 0.05%
Rise of the Tomb Raider™ (50P) - lvl 0 / ~24h
Before
7420 / 0.01%
After
5937 / 0.02%
Even though there are not enough data to prove anything, it's possible to infer that the odds seem to have, at least, doubled in most cases.
1 hour AAA GAs seem the most unaffected by the new system, while 1 hour bundled GAs seem to be the most affected ones.
I can't say if it's better or worse than the previous system yet, for 2 reasons:
First, I'd need to know the amount of points we were getting every 15 minutes, on average, before the update to compare with the amount we're getting now.
Second, I'd have to compare how both quantities translate into winning GAs since the odds have changed a lot and there aren't enough finished GAs to do it yet.
TL;DR: Wait at least a month or two before you start being apocalyptic.
What do you think translate to winning more GAs: More points per hour or better odds?
Comment has been collapsed.