Jump to content


Jeers for Personal Reserves

Personal Reserves

  • Please log in to reply
41 replies to this topic

Ape_Drape #21 Posted Jul 26 2017 - 15:00

    Captain

  • Players
  • 37233 battles
  • 1,347
  • [SOT] SOT
  • Member since:
    06-13-2011

View PostDemonic_Angel_of_Death, on Jul 25 2017 - 23:31, said:

The crew XP coupled with an XP reserve comes in handy, but the credits and free XP ones really aren't worth it...

 

Hey now... those credit PRs help me make the silver per match I used to earn in my premium tanks before PRs came out.

Chalybos #22 Posted Jul 26 2017 - 15:13

    Staff sergeant

  • Players
  • 20830 battles
  • 467
  • Member since:
    04-18-2011

Last night, from about 8-11 EST, I had 7 losses in a row, then 7 wins in a row, 2 losses, then 6 wins.  All while running boosters.  All that proves is that my game was streaky last night.

 



Darkbee2Bee #23 Posted Jul 26 2017 - 15:16

    Major

  • Players
  • 31459 battles
  • 2,925
  • [-FG-] -FG-
  • Member since:
    06-23-2013

I think true randomness is hard for humans to comprehend.  We like patterns. We want to see patterns.  Intuitively, when we toss a coin we want to see a fairly, nice, neat uniform, heads, tails, heads, tails.  We might throw in the the occasional double result, heads, tails, tails, heads, tails, heads, heads.. and call it random. That is random to us.. still neat and tidy but not 100% predictable.  The truth is, that's not random at all.  In a true random scenario, a series of all heads or all tails is possible and moreover, given a large enough sample size it WILL happen.

 

As far as losses being generated from running personal reserves, alongside just an unfortunate random streak, is it possible that some psychology is at play?  Because you've enabled those personal reserves you want to perform better to maximize their effectiveness.  However, your first game doesn't go according to plan so you try a little harder, play a little more aggressively in the next match and over do it.  Now you begin chasing your tail and it's just a downward spiral from there, trying increasingly harder to make up for all the perceived lost bonuses.

 

I haven't personally noticed any correlation between winning/losing streaks and personal reserves.  I love personal reserves, I think it's a great way to shorten grinds and maximize other bonuses available (such as times 3 and times 5 events).  However, if you feel strongly about it then why not lay off them a while, seem if things turn around and then maybe fire them up another time.



Nunya_000 #24 Posted Jul 26 2017 - 16:17

    Major

  • Players
  • 20683 battles
  • 9,792
  • [PACNW] PACNW
  • Member since:
    09-20-2013

View PostMartian_Sentry, on Jul 25 2017 - 17:44, said:

We do KNOW there are a number of elements operating in MM to position each player within a match, INCLUDING an overall attempt to keep the bell curve as narrowly close to the 50% point as possible.  

 

How do we know this???  If you are referring to the patent, it does not say anything close to that.  We also do not know if all items explained in the patent are "operating" in MM.

 

After going to great lengths to explain how "logical" you are, you forgot the most important question of logic.  That question is "WHY".  Why would WG go to the expense of creating a reward, that are very easy to acquire and do not really provide a huge amount of benefit, just to use those rewards to screw us in future battles?  A logical approach would be to create rewards gain by playing the game and then have those rewards actually help player enjoy the game so that they continue to play...and earn more rewards (a vicious cycle).  Players that feel that they are being screwed do not play the game for long.  People do not spend money on a game they do not play.

 

You claim that this cycle is easily observed.  A logical approach would be to make any manipulation subtle so that it is not easily observed.

 

You also claim that MM is "rigged" to keep us as close to a 50% WR as possible, but then claim that these PR cause losing streaks.  That seems fairly contradictory.

 

The outcomes of games are mainly dictated on how well 15 players perform in any given battle.  Our individual performances in battles are not static, they are dynamic.  A good player can make poor decisions in an individual battle and have a bad game.  A poor player can make good decisions in a specific battle and have a good game.  How well all 15 players perform in a battle is not something that can be rigged.



Nunya_000 #25 Posted Jul 26 2017 - 16:29

    Major

  • Players
  • 20683 battles
  • 9,792
  • [PACNW] PACNW
  • Member since:
    09-20-2013

View PostDarkbee2Bee, on Jul 26 2017 - 06:16, said:

I think true randomness is hard for humans to comprehend.  We like patterns. We want to see patterns. 

 

There is even a word for it.  It is called "Apophenia".  Some behavioral scientists even believe that it is an indication of early signs of possible schizophrenia in those people that are strongly convinced these patterns exist even though there is little in the way of proof outside of personal observations.  This trait is often seen in habitual gamblers that see patterns in random events such as "which card will be turned over next", or "what number will the ball fall on" based on some perceived observed pattern.



Weapon_of_Proliferation #26 Posted Jul 26 2017 - 17:44

    Staff sergeant

  • Players
  • 25336 battles
  • 481
  • [G-M-U] G-M-U
  • Member since:
    12-25-2011

View PostMartian_Sentry, on Jul 26 2017 - 01:44, said:

First - I appreciate the honest and straight forward replies.  I wasn't trying to start a flame war but simply state my observations and to gauge, from the replies, how others felt about PR's.

 

In reply though ---

 

I am aware of confirmation bias, but that does not rule out the validity of observation.  If it did science wouldn't make any progress at all.  Its the job of statistics to validate the significance of those observations.  So its not proper to simply dismiss my observations out-of-hand as automatically CB.  Besides, it would be logical to presume as a starting point that PR's are a VALUABLE 

commodity and so the CB should be in the FAVOR of PR's.  And that would make evidence/observation to the contrary not CB.

 

 

As I stated in the original post, you cannot know this unless you have explicit knowledge of the algorithm underlying the MM.  You may or want to believe it but you cannot state it categorically 

as you've done.  We do KNOW there are a number of elements operating in MM to position each player within a match, INCLUDING an overall attempt to keep the bell curve as narrowly close

to the 50% point as possible.  So its certainly a possibility the MM algorithm notes the activation of PR's AND creates a bias that affects that player's placement in a match.

 

 

I'll point again to my comments about Conf. Bias.  It's logical to assume at the beginning that PR's ARE a good thing so that is what "I Think".  It was only experience and observation that pushed

AGAINST what I initially thought and has begun to CHANGE my opinion.  That is hardly a self-fulfilling prophecy.

 

 

Why do you assume I'm not perfectly aware that correlation does not equate to causation?  But again it is the job of multiple observations and statistics to establish or exclude the link.

Remember the old adage - Once Is Chance, Twice is Coincidence, Third Time Is Enemy Action.

So, if you do keep eating ham sandwiches and then later continue to be stung by a bee - well, there MAY be something in those sandwiches attracting the bees to you.

 

Again, thanks for all your replies.  I appreciate the effort.

 

This is what happens when you keep the fluoride intake to a minimum.

 

+1



Uilleam72 #27 Posted Jul 26 2017 - 19:15

    Sergeant

  • -Players-
  • 12888 battles
  • 119
  • Member since:
    03-18-2016

What I'd like to see happen with PR is the ability to trade them in or exchange them for ones that are more useful. Some of us don't have 4 hours to play or do not want to play for 4 hours straight. I rarely have the time to play for 4 hours. I find sleep more productive than getting stuck in a 2 hour losing streak because you want to maximize your PR or you hope the next team isn't going to suck. Some days I don't care how much time I have left on PR if I'm on a losing streak and just log off.

 



Martian_Sentry #28 Posted Jul 26 2017 - 19:17

    Sergeant

  • Players
  • 26606 battles
  • 165
  • Member since:
    06-26-2011

View PostNunya_000, on Jul 26 2017 - 15:17, said:

 

How do we know this???  If you are referring to the patent, it does not say anything close to that.  We also do not know if all items explained in the patent are "operating" in MM.

 

After going to great lengths to explain how "logical" you are, you forgot the most important question of logic.  That question is "WHY".  Why would WG go to the expense of creating a reward, that are very easy to acquire and do not really provide a huge amount of benefit, just to use those rewards to screw us in future battles?  A logical approach would be to create rewards gain by playing the game and then have those rewards actually help player enjoy the game so that they continue to play...and earn more rewards (a vicious cycle).  Players that feel that they are being screwed do not play the game for long.  People do not spend money on a game they do not play.

 

You claim that this cycle is easily observed.  A logical approach would be to make any manipulation subtle so that it is not easily observed.

 

You also claim that MM is "rigged" to keep us as close to a 50% WR as possible, but then claim that these PR cause losing streaks.  That seems fairly contradictory.

 

The outcomes of games are mainly dictated on how well 15 players perform in any given battle.  Our individual performances in battles are not static, they are dynamic.  A good player can make poor decisions in an individual battle and have a bad game.  A poor player can make good decisions in a specific battle and have a good game.  How well all 15 players perform in a battle is not something that can be rigged.

 

Yes, I was referring, among other things to the patent.  And while I haven't read the entire document parts that I have certainly point in the direction of a constrained distribution

around the 50% mark.  Also, there have been previous forum posts on this topic showing the WoT's distribution curve based on actual player data and those curves have been 

much more of a spike (centered @ 50%) than a more standard bell curve.  And whether ALL of the elements in the patent are expressed in MM is not at issue; the only point

being made is the MM algorithm does factor in numerous elements so the activation of a PR COULD be one of them.

 

I would strongly disagree that a "why" i.e. motivation is needed in a logical deduction.  We can conduct a murder investigation, collecting a chain of hard evidence, and clearly

deduce that Smith is the murderer without ever knowing WHY he did it.  I'm sure any prosecutor WANTS to give the jury a motivation as part of the package but its not necessary

to prove the case.  As to a POSSIBLE motivation in WOT, it could easily be the "Dr. Feelgood" affect I mentions - people LIKE receiving "free stuff".  But to mitigate the loss to

their side of the ledger the MM COULD bias match placement just a bit.  Remember - roulette wheels have a Zero and Double Zero to insure the House Only wins sometimes.

 

And a point to remember - this is ostensibly a "free to play" game, so its fair to presume people WILL put up with more perceived "unfairness" than they otherwise might.

 

As to the cycle being subtle or not:  simply recall all the discussion over the last few patches about an "improved" MM.  WG stated repeatedly a primary constraint on any 

changes to MM was keeping the wait time as short as possible.  Obviously convenience overrides subtly.  Also I would expect WG to depend on just the skepticism you're

expounding to blunt most players suspicions about such a PR bias taking place.

 

I never used the word "rigged". We both know that word is highly prejudicial here in the forums and I avoided it both in form and function.  The word I used was "programmed"

and that is an accurate description of MM.  You seem to be making the assumption that a bias (if it exists) MUST ALWAYS be enforced regardless of a player's win rate.

Why ?  If MM is indeed attempting to constrain the population around the 50% mark it would certainly make sense to LESSEN or even eliminate the PR bias the more

negative (from the 50% mark) the player's win rate.  Therefore there is no necessary contradiction.

 

In your last paragraph I completely agree with your first four sentences but take exception to the last sentence (and suggest again you reframe from the word "rigged"

due to its inflammatory nature).  No, the MM certainly can't predict how any individual player will perform in any single battle, but a poorly engineered MM CAN certainly

create an unbalanced team distribution, leading to a very unbalanced outcome.  And the proof is right in front of us, for why else would have WG gone to all the trouble

of creating a heavily revised MM  with the template and other systems to provide a better balance if in fact the MM could not effect the outcome of the battle ?

 



164thLeichteAfrikaDiv #29 Posted Jul 26 2017 - 19:21

    Corporal

  • -Players-
  • 3593 battles
  • 87
  • [GSH] GSH
  • Member since:
    05-10-2017
We have a joke in our clan about Personal Reserves. If you want to work on your losing rate, use them. Sometimes you win, but most of the time you lose more than you usually do. Last night I played a reserve and lost 9 straight games. i quit playing after that.

Edited by 164thLeichteAfrikaDiv, Jul 26 2017 - 19:22.


FrozenKemp #30 Posted Jul 26 2017 - 19:44

    Major

  • Players
  • 42110 battles
  • 4,271
  • Member since:
    04-24-2011

I have been playing almost all of my game game time for the last 8+ months with an XP personal reserve activated, and shocking it doesn't have any perceptible negative effect.

 

Try playing more than an hour before jumping to conclusions.  And I mean that - if you have reserves hoarded, just stop hoarding.  Use them!  Play a whole week, always with reserves.  See how things are. 


Edited by FrozenKemp, Jul 26 2017 - 19:46.


Nunya_000 #31 Posted Jul 26 2017 - 19:52

    Major

  • Players
  • 20683 battles
  • 9,792
  • [PACNW] PACNW
  • Member since:
    09-20-2013

View PostMartian_Sentry, on Jul 26 2017 - 10:17, said:

 

Yes, I was referring, among other things to the patent.  And while I haven't read the entire document parts that I have certainly point in the direction of a constrained distribution

around the 50% mark.  Also, there have been previous forum posts on this topic showing the WoT's distribution curve based on actual player data and those curves have been 

much more of a spike (centered @ 50%) than a more standard bell curve.  And whether ALL of the elements in the patent are expressed in MM is not at issue; the only point

being made is the MM algorithm does factor in numerous elements so the activation of a PR COULD be one of them.

 

I have read the whole patent.  The only inference I can remember is the statement that it may place players more often as low tier if they have been winning (more of a challenge) and high tier if they have been losing (easier).  There is nothing that states players are placed on poor teams or good teams in order to keep their WR close to 50% as possible.  Whether a player is low tier or high tier does not dictate their chance to win or lose.  In fact it could have the opposite effect...and even more so with the new MM.  A poor players at high tier could actually decrease a team's chance of winning....while a good player at low tier would not necessarily decrease the win/loss chance.

 

There is no such thing as a 'standard bell curve'.  EVERY bell curve has it's own shape based on what is measured.  Some bell curves are not even symmetrical.

 

Yes, it is certainly POSSIBLE that MM is "programmed" to keep players as close to possible to a 50% WR.  I do not think it is probable.  Why go to the expense to program something that will happen naturally in a random format.

 

Block Quote

I would strongly disagree that a "why" i.e. motivation is needed in a logical deduction.  We can conduct a murder investigation, collecting a chain of hard evidence, and clearly

deduce that Smith is the murderer without ever knowing WHY he did it.  I'm sure any prosecutor WANTS to give the jury a motivation as part of the package but its not necessary

to prove the case.  As to a POSSIBLE motivation in WOT, it could easily be the "Dr. Feelgood" affect I mentions - people LIKE receiving "free stuff".  But to mitigate the loss to

their side of the ledger the MM COULD bias match placement just a bit.  Remember - roulette wheels have a Zero and Double Zero to insure the House Only wins sometimes.

 

In murder cases, there are 3 things that are ALWAYS looked at to determine guilt: Means, Motive, and Opportunity.  That does not mean that the motive is always obvious, but it is always a major item considered.  The WHY is always a consideration in logical thinking.  Just because something is "possible" that does not mean it is "practical" when the WHY is considered.

 

Block Quote

And a point to remember - this is ostensibly a "free to play" game, so its fair to presume people WILL put up with more perceived "unfairness" than they otherwise might.

 

Some people will put up with some "unfairness"....to a point.  Most of us play for entertainment.  When something no longer entertains us, we will move on to other things.  The feeling of being manipulated quickly creates frustrations (see your OP).  Most sane frustrated players desire to end their frustration quickly (i.e. your solution was to no longer use PRs).

 

Block Quote

As to the cycle being subtle or not:  simply recall all the discussion over the last few patches about an "improved" MM.  WG stated repeatedly a primary constraint on any 

changes to MM was keeping the wait time as short as possible.  Obviously convenience overrides subtly.  Also I would expect WG to depend on just the skepticism you're

expounding to blunt most players suspicions about such a PR bias taking place.

 

Which brings us back to the WHY.  If queue times are important to be kept as short as possible, the idea that MM is "programmed" to manipulate our WRs would be in conflict with that.  Why program something that increases queue times and produces the same results that a non-"programmed" MM would create in a random format.  If they could "program" MM to manipulate our WRs, they could certainly "program" it to be subtle so that streaks do not happen with little differences to queue times.

 

Block Quote

I never used the word "rigged". We both know that word is highly prejudicial here in the forums and I avoided it both in form and function.  The word I used was "programmed"

and that is an accurate description of MM.  You seem to be making the assumption that a bias (if it exists) MUST ALWAYS be enforced regardless of a player's win rate.

Why ?  If MM is indeed attempting to constrain the population around the 50% mark it would certainly make sense to LESSEN or even eliminate the PR bias the more

negative (from the 50% mark) the player's win rate.  Therefore there is no necessary contradiction.

 

An argument of semantics.  If the game is "rigged" to keep our WRs close to 50% as possible and/or PRs make you lose more, then it was certainly "programmed".....and if there is programming to make up lose more when we activate PRs or to keep our WRs at 50%, then this game is certainly "rigged".

 

Block Quote

In your last paragraph I completely agree with your first four sentences but take exception to the last sentence (and suggest again you reframe from the word "rigged"

due to its inflammatory nature).  No, the MM certainly can't predict how any individual player will perform in any single battle, but a poorly engineered MM CAN certainly

create an unbalanced team distribution, leading to a very unbalanced outcome.  

 

Yes, some battles are unbalanced.  That would be normal in a random format.  In fact, it is pretty much the definition of random.

 

Block Quote

 And the proof is right in front of us, for why else would have WG gone to all the trouble

of creating a heavily revised MM  with the template and other systems to provide a better balance if in fact the MM could not effect the outcome of the battle ?

 

There is absolutely no logic in the above sentence.  MM sets up the teams.  It does not dictate the outcome.  Certainly, a severely unbalanced in teams can be predicted on what the outcome would LIKELY be.  However, the players still need to fight the battle.

 

The fallacy you present is implying that the motivation to change the MM was because they know it can affect the outcomes.  There are numerous reasons for the changes.  Players have been complaining for years about one team having more higher tier tanks than the other....or that one team had more TDs...or that there were too many higher tier tanks a person had to fight against when they were low tier.  This change address all of these issues, but created other issues with it's application.  You are defining WG motives for the change based on your own bias assumptions.



Darkbee2Bee #32 Posted Jul 26 2017 - 19:54

    Major

  • Players
  • 31459 battles
  • 2,925
  • [-FG-] -FG-
  • Member since:
    06-23-2013
@Martian_Sentry... just wanted to give you some props for making an actual debating post rather than a monkey poop flinging one.  Seems pretty rare these days. I don't agree with your points and I'd love to respond to every one of them but I don't have the time.  I will say though that your murder analogy is kind of flawed in that, in that case we "know" a crime has been committed.  In the case of WG and rigging, we don't know for sure; there is no "body" (without a body no crime has been committed!).  In which case, all we're left with is speculation and thus the "why", so that we can try to formulate hypotheses to test.  So I'd would argue that the "why" is actually very important in this instance.

vinnybagadonuts #33 Posted Jul 26 2017 - 21:18

    Staff sergeant

  • Players
  • 25976 battles
  • 398
  • [UBH-T] UBH-T
  • Member since:
    06-14-2011

look at it another way (since math isnt a strong suit)...

 

even in losses, with a 100% xp booster...you get the same XP youd get in a win.



Blacksheep_14 #34 Posted Jul 26 2017 - 22:00

    Corporal

  • -Players-
  • 16981 battles
  • 50
  • [GTTDS] GTTDS
  • Member since:
    05-13-2014
Anytime I activate a PR I am assured of being bottom tier 95% of the time, as opposed to 90% of the time if not using PR. Plus  MM makes sure you are on a team full of players that lemming to one side of the map.

120mm_he #35 Posted Jul 26 2017 - 22:01

    Major

  • Players
  • 13693 battles
  • 15,303
  • [PBS] PBS
  • Member since:
    02-17-2011
My experience with pr is that you get a slight shift in mm bias and are more often put on baddie filled teams. This bears out by me observing the xvm stats of my team when using pr and not using pr. It's not a huge shift but consistently getting half your team with sub 40% win rates does drag it down at times. However the boosts always nets me significantly more overall xp even if the sessions win rate was lower because the wins really float out the totals. With all three activated along with a crew xp mission I can get 10k+ xp on the crew on a good win and not much less on an average game. So its a choice between crew and tank xp generation or your overall win rate. 

BillT #36 Posted Jul 26 2017 - 23:25

    Captain

  • Beta Testers
  • 23848 battles
  • 1,636
  • [FADES] FADES
  • Member since:
    08-13-2010

View PostMartian_Sentry, on Jul 25 2017 - 17:27, said:

But TEN in a row is definitely NOT normal

 


Statistically, 10 losses in a row ARE normal.  It's an expected result from a normal distribution. If your expected win rate is 49% (average), then every time you start a game, there is a 0.08% chance that it will be the beginning of a 10-game losing streak.  That's not high, but you've played 25,000 battles, so you've had 25000 chances for a 10-game losing streak to begin. Statistically, you should have had about 20 such losing streaks by now.

Yes, the game COULD be rigged to punish you for using personal reserves.  And George W. Bush COULD have blown up the World Trade Center. What both theories lack is a credible motive that explains why someone would perpetrate such a thing.  How does WG profit from designing the matchmaker to punish (and thus, frustrate) players who use personal reserves?  Heck, they SELL those things... they want people to buy them.



BillT #37 Posted Jul 26 2017 - 23:29

    Captain

  • Beta Testers
  • 23848 battles
  • 1,636
  • [FADES] FADES
  • Member since:
    08-13-2010

View PostLord_Hiney, on Jul 25 2017 - 19:47, said:

Today I ate a ham sandwich, then later I was stung by a bee. Last time I was stung by a bee, I had also eaten a ham sandwich.

Guess I better stop eating ham sandwiches.

 

Naw.  Just stay inside (and play WOT) after you eat one.  Doctor's orders.

Kliphie #38 Posted Jul 26 2017 - 23:38

    Captain

  • Players
  • 27625 battles
  • 1,436
  • [IOC] IOC
  • Member since:
    07-20-2012

View PostBillT, on Jul 26 2017 - 17:29, said:

 

Naw.  Just stay inside (and play WOT) after you eat one.  Doctor's orders.

 

The BEEES .... are IN ... the HOUSE!!!!!!!!!!!

BillT #39 Posted Jul 26 2017 - 23:51

    Captain

  • Beta Testers
  • 23848 battles
  • 1,636
  • [FADES] FADES
  • Member since:
    08-13-2010

View PostDarkbee2Bee, on Jul 26 2017 - 09:16, said:

I think true randomness is hard for humans to comprehend.  We like patterns. We want to see patterns.  Intuitively, when we toss a coin we want to see a fairly, nice, neat uniform, heads, tails, heads, tails.  We might throw in the the occasional double result, heads, tails, tails, heads, tails, heads, heads.. and call it random. That is random to us.. still neat and tidy but not 100% predictable.  The truth is, that's not random at all.  In a true random scenario, a series of all heads or all tails is possible and moreover, given a large enough sample size it WILL happen.

 

 

Spot on.  As a visual demo, the reader is invited to guess which of the following strings of twelve coin toss results was randomly generated:

(H=heads, T=tails)

 

A)  T T T T T T H H T T H H

B)  T H T H H T T H H T H T

C) H T H T H T H T H T H T

 

Since you are all expecting a trick, it'll be no surprise when I tell you A is the random one . What *might* surprise you is that result A) was what I got on just my second attempt to generate 12 coin tosses.  It wasn't hard to get, even though it's an extreme result.  The odds of generating the exact sequence of string A are only 0.0244%.  And that's exactly the same as the odds of generating string B or C.

 

But A and C stand out to you because you can spot patterns.  Six in a row. A repeating sequence.  Your mind evolved to suspect some kind of conscious action behind such patterns, something you can take advantage of.  String B has an equal number of H and T and no obvious pattern, so it looks random.



BillT #40 Posted Jul 26 2017 - 23:57

    Captain

  • Beta Testers
  • 23848 battles
  • 1,636
  • [FADES] FADES
  • Member since:
    08-13-2010

View PostMartian_Sentry, on Jul 25 2017 - 20:44, said:

 

Remember the old adage - Once Is Chance, Twice is Coincidence, Third Time Is Enemy Action.

 

 

It's not *that* old.  Ian Fleming wrote it for James Bond in "Goldfinger".  (He used "happenstance,",not "chance".) So it's not really an adage, either. 

 

But to your point... you described one bad night of matchmaking while using PRs.  That's happenstance.  If you get another 10-game losing streak the next time you use PRs, you'll be up to coincidence.  Document three such cases in a row, and I'll be pretty convinced you're correct about the MM.







Also tagged with Personal Reserves

1 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users