I looked at Tempostorm's meta snapshot just to see what's going on with the meta. When I looked at their new snapshot today, I was a little confused and wondered what they were thinking. They placed control warlock at tier 1 as the #1 deck and moved down odd paladin to tier 2. If I go to HSReplay, I see odd paladin with a 57.21% winrate and control warlock with less than 50%. Am I missing something? Is Tempostorm just unreliable for meta snapshots anymore? What do you think?
I used to use Tempostorm's meta snapshot just to see what's going on with the meta. When I looked at their new snapshot today, I was a little confused and actually upset. They placed control warlock at tier 1 as the #1 deck and moved down odd paladin to tier 2. If I go to HSReplay, I see odd paladin with a 57.21% winrate and control warlock with less than 50%. Am I missing something? Is Tempostorm just unreliable for meta snapshots anymore? What do you think?
I have no idea how tempostorm makes their list, but it's possible that they use something more involved than just win rate. Maybe they look at popularity too. Maybe they look at win rate at certain levels? Pally could be 57.21% across the board with the percentage being skewed by higher success at level 15 and lower at level 5?
Again, I don't know. Just spitballing. Either way, you don't need a website to tell you Control Warlock and Aggrodin are very good decks.
hsreplay is also not perfect as harder deck are played by people at lower ranks and thus winrate goes down, however, obviously it gives a general idea. It is weird actually to see that tempo storm snapshot is so out of touch with the meta.
I would consider the vicioussyndicate.com report the gold standard.
But that doesn't make the other sources worthless.
-Tempostorm is the opinion of a small group of pro players. So is more relevant if you are pushing for tournament success. Less good for an average player on ladder.
-HSreplay gets a very large amount of data, but their statistical model is suspect. Good for looking at muligan data and for early data following an expansion or patch.
Vicious syndicate is a nice alternative... and they didn’t even miss the best warrior archetype of warrior (quest warrior without Baku version) unlike tempostorm. Though arguably what tempo storm is assessing is tournament meta and not ladder.
Tempo storm is reliable, hsreplay only looks at win rate while tempo storm takes a week and half of in depth analysis to form their meta snapshot
Hsreplay have all the data from hsdecktracker and u can filter by legend or what u want with premium by far hsreplay is best site for looking best decks and info about meta
Most people here saying Tempostorm is trash. Their meta snapshot shows which decks are the best but also quite popular. Odd paladin is popular but at lower ranks since it's garbage in legend against the field
It may be a issue with the lower ranks, their meta snapshot shows exactly the decks I'm facing in legend 1000-3000
Tempo storm is reliable, hsreplay only looks at win rate while tempo storm takes a week and half of in depth analysis to form their meta snapshot
Hsreplay have all the data from hsdecktracker and u can filter by legend or what u want with premium by far hsreplay is best site for looking best decks and info about meta
Problem with hsreplay that if not enough cards were played by opponent to determine archetype then it gets classified as “other” so winrates of the decks are probably lower then what they state.
vicious syndicate for example disregards all the games where they were not able to determine archetype of the opponent deck. that makes winrates they post lower but very reliable and mostly it lists winrates against top meta decks which is arguably the most relevant statistics anyway.
Odd pally did really good the first few days, but the more it is played the easier it is to counter. Odd hunter had a similar scenario, but is even worse.
Tempo storm is reliable, hsreplay only looks at win rate while tempo storm takes a week and half of in depth analysis to form their meta snapshot
Hsreplay have all the data from hsdecktracker and u can filter by legend or what u want with premium by far hsreplay is best site for looking best decks and info about meta
Problem with hsreplay that if not enough cards were played by opponent to determine archetype then it gets classified as “other” so winrates of the decks are probably lower then what they state.
vicious syndicate for example disregards all the games where they were not able to determine archetype of the opponent deck. that makes winrates they post lower but very reliable and mostly it lists winrates against top meta decks which is arguably the most relevant statistics anyway.
Vs have 130k games in all ranks it not enough data Hsreplay have more games in 1 day
In Hsreplay u can analize meta of the day last meta snapshot in tempostorm control warlock was tier 3 ridicule
Tempo storm is reliable, hsreplay only looks at win rate while tempo storm takes a week and half of in depth analysis to form their meta snapshot
Hsreplay have all the data from hsdecktracker and u can filter by legend or what u want with premium by far hsreplay is best site for looking best decks and info about meta
Problem with hsreplay that if not enough cards were played by opponent to determine archetype then it gets classified as “other” so winrates of the decks are probably lower then what they state.
vicious syndicate for example disregards all the games where they were not able to determine archetype of the opponent deck. that makes winrates they post lower but very reliable and mostly it lists winrates against top meta decks which is arguably the most relevant statistics anyway.
Vs have 130k games in all ranks it not enough data Hsreplay have more games in 1 day
In Hsreplay u can analize meta of the day last meta snapshot in tempostorm control warlock was tier 3 ridicule
Yes to small of a data set is a problem that VS has... but you overall stats on HSreplay should be taken with a grain of salt like overall winrate, unfortunaly a lot of losses of Murloc Paladin for example will go into "Other Paladin" because guy never drew a single Murloc and lost... the only proper way of analyzing decks/archetypes on HSReplay is looking at "Matchups" and even then some wins/loses will be missing from statistics because of other Category... Now if there was some way to exclude All "Other [name of class]" from statistics on HSReplay it would be interesting to see how it would work.
P.S. actually right... in global match ups table you can exclude stuff... Totally forgot... pictures an interesting look on the meta and does drop Control warlock all the way to the bottom of the Tier 3. (Basically control lock is bad against what is popular in the meta)
Both analyses are fair, you just have to look at their context.
TempoStorm uses a holistic approach with Data, Pro Opinions, Forecasting, etc. to come up with a list aimed at the highest levels of play. In their report, yes Odd Paladin is Tier 2 because of prominent control decks like Warlock being much more piloted at the higher level due to the ability to influence their win rates. Lists like these don't look to tell what's going on in the meta with stats, they tend to tell you that if Odd Paladin is played a lot, their counters, namely Control Lock in this will be picked more often and given the environment has a better chance at surviving the ladder.
HSReplay or ViciousSyndicate are pure data aggregators. They tell you what's going on currently (Without Context) based on incoming data. According to that, at all levels of play, Odd Paladin is seeing a 57.2% win rate and they can tell you what that matchup is like against another deck based on the data. These types of analyses are very good at a rudimentary look at the meta, and helpful for players would need a broad strategy overview and not such a refined look like what TempoStorm is aiming to bring.
Overall, I would probably say that a list like TempoStorm is geared towards maybe 10-15% of this site, whereas ViciousSyndicate can provide helpful insight to a majority of the same audience as you'll find here on HearthPwn. That doesn't mean one is objectively better or worse than another, it just means they provide different contexts, and whichever one applies to you more is the one you'll favor.
Well, I certainly understand what TempoStorm is trying to do; I just think they're doing it poorly.
In soccer/football, you want to be where the ball is, not were the ball was. When it comes to planning ahead, that means you don't want to go to where the ball is now -- you want to go to where the ball will be. Data aggregators like HSReplay and ViciousSyndicate can give very detailed information about where "the ball" is now, because that's objective; what they don't offer is subjective analysis of where "the ball" will be.
Thus a site like TempoStorm, which will never beat HSReplay or even Vicious Syndicate on data, tries to predict the future. In this case that means identifying not the winning decks of today, but the deck (or decks) that will best counter the winning decks of today.
The problem I see with TempoStorm's most recent Meta Snapshot isn't that it tries to make daring predictions, but it deviates from matchup data (without sufficient cause). They claim that "Control Warlock really only has one true enemy right now on ladder, and that's Quest Rogue," but that's not accurate; Control Warlock has an unfavorable matchup against Cubelock, and this problem is worse at high ranks than at low ranks. Yet inexplicably they list Control Warlock as favored against Cubelock, despite a rather standard decklist and no expectation -- or more importantly, explanation -- of a particularly winning piloting technique. Not to mention bad matchups they chart against Spiteful Druid, Control Priest, Taunt Druid and Tempo Mage.
In my recent Standard meta predictions thread (https://www.hearthpwn.com/forums/hearthstone-game-modes/standard-format/217154-i-have-modeled-58-archetypes-heres-the-future), I also tried to predict who the Paladin Killer would be, but my answer was Control Priest due to its favored matchups against both Cubelock and Even Paladin. If I was managing Tempo Storm, I probably would have listed it as the #2 or #3 deck for next week... behind Cubelock, because it's simply not reasonable to assume it will be dethroned any time soon.
But to get an empirically 42% matchup as a 60% matchup... well that's strangely incompetent. I almost wonder if they're deliberately giving bad information.
control warlock is tier 1 in legend, people saying tempostorm is wrong are probably rank 20
Control warlock is really good at legend and at rank 4 or below it is misplayed so heavily that it skews the stats... Still cubelock is better just because doomguard highroll potential makes bad match ups not as bad... Also definition of tier 1 is kinda vague... But control lock is certainly over 50% winrate if played correctly, but that requires high skill and good knowledge of match ups.
control warlock is tier 1 in legend, people saying tempostorm is wrong are probably rank 20
Control warlock is really good at legend and at rank 4 or below it is misplayed so heavily that it skews the stats... Still cubelock is better just because doomguard highroll potential makes bad match ups not as bad... Also definition of tier 1 is kinda vague... But control lock is certainly over 50% winrate if played correctly, but that requires high skill and good knowledge of match ups.
That suggests that Control is the proper deck to put at the top though. If the issue with Control Warlock is that it's harder to use while the benefit of cubelock is that it highrolls to cover for bad matchups, that essentially means you are trading greater reliability for the ability to gamble-win a few extra times.
This sort of situation shows up a lot actually in hearthstone. Key examples I remember are the Miracle Rogue meta in 2014 and the Patron meta. In both cases the deck proved to be easy to overwhelm at the lower ranks as crafty opponents can exploit weaknesses caused by hard to spot misplays. Thus for the general public running a faster, easier deck (face hunter in 2014 or secret paladin in 2015) was better and provided a higher win rate.
But that doesn't mean Miracle or Patron were lower tier than Face/Secret. In the end the former decks were the stronger which is why tournaments and high legend worshiped them. It just showed that what worked for the pro player wouldn't work for the casual/semi-competitive.
The problem isn't Tempostorm or VS. It's this odd assumption we are making that 100% of the game is based on what deck you run and that the best deck for a weak player is the same as the best deck for the pro player.
As stated earlier, Tempostorm is meant for the high end pros. If you aren't pushing high legend or running tournaments then its going to be inconsistent with your current skill level and environment.
VS and HSreplays are skewed towards the entire ladder, which mostly consists of lower skilled and casual environments. If you ARE seeking the pro scene then these sites will be misleading to you as their datasets don't account for skill level or the potential in the deck when played optimally.
It's honestly awesome that we have both sets of data available as most games tend to carry just one or the other. We just need to realize our fortune and use each part of the data for how they are meant to be used.
I looked at Tempostorm's meta snapshot just to see what's going on with the meta. When I looked at their new snapshot today, I was a little confused and wondered what they were thinking. They placed control warlock at tier 1 as the #1 deck and moved down odd paladin to tier 2. If I go to HSReplay, I see odd paladin with a 57.21% winrate and control warlock with less than 50%. Am I missing something? Is Tempostorm just unreliable for meta snapshots anymore? What do you think?
hsreplay is also not perfect as harder deck are played by people at lower ranks and thus winrate goes down, however, obviously it gives a general idea. It is weird actually to see that tempo storm snapshot is so out of touch with the meta.
According to them, zoo is teir4, in reality, it's a solid tier2 deck.
Tempostorm is hot trash
I would consider the vicioussyndicate.com report the gold standard.
But that doesn't make the other sources worthless.
-Tempostorm is the opinion of a small group of pro players. So is more relevant if you are pushing for tournament success. Less good for an average player on ladder.
-HSreplay gets a very large amount of data, but their statistical model is suspect. Good for looking at muligan data and for early data following an expansion or patch.
Tempostorm was never reliable.
Tempo storm is reliable, hsreplay only looks at win rate while tempo storm takes a week and half of in depth analysis to form their meta snapshot
Check out my Earth, Wind, and Fire - Elemental Shaman deck!
Vicious syndicate is a nice alternative... and they didn’t even miss the best warrior archetype of warrior (quest warrior without Baku version) unlike tempostorm. Though arguably what tempo storm is assessing is tournament meta and not ladder.
I am envoy from nowhere in nowhere. Nobody and nothing have sent me. And though it is impossible I exist. ©Trimutius
Most people here saying Tempostorm is trash. Their meta snapshot shows which decks are the best but also quite popular. Odd paladin is popular but at lower ranks since it's garbage in legend against the field
It may be a issue with the lower ranks, their meta snapshot shows exactly the decks I'm facing in legend 1000-3000
I am envoy from nowhere in nowhere. Nobody and nothing have sent me. And though it is impossible I exist. ©Trimutius
Odd pally did really good the first few days, but the more it is played the easier it is to counter. Odd hunter had a similar scenario, but is even worse.
I am envoy from nowhere in nowhere. Nobody and nothing have sent me. And though it is impossible I exist. ©Trimutius
Both analyses are fair, you just have to look at their context.
TempoStorm uses a holistic approach with Data, Pro Opinions, Forecasting, etc. to come up with a list aimed at the highest levels of play. In their report, yes Odd Paladin is Tier 2 because of prominent control decks like Warlock being much more piloted at the higher level due to the ability to influence their win rates. Lists like these don't look to tell what's going on in the meta with stats, they tend to tell you that if Odd Paladin is played a lot, their counters, namely Control Lock in this will be picked more often and given the environment has a better chance at surviving the ladder.
HSReplay or ViciousSyndicate are pure data aggregators. They tell you what's going on currently (Without Context) based on incoming data. According to that, at all levels of play, Odd Paladin is seeing a 57.2% win rate and they can tell you what that matchup is like against another deck based on the data. These types of analyses are very good at a rudimentary look at the meta, and helpful for players would need a broad strategy overview and not such a refined look like what TempoStorm is aiming to bring.
Overall, I would probably say that a list like TempoStorm is geared towards maybe 10-15% of this site, whereas ViciousSyndicate can provide helpful insight to a majority of the same audience as you'll find here on HearthPwn. That doesn't mean one is objectively better or worse than another, it just means they provide different contexts, and whichever one applies to you more is the one you'll favor.
Well, I certainly understand what TempoStorm is trying to do; I just think they're doing it poorly.
In soccer/football, you want to be where the ball is, not were the ball was. When it comes to planning ahead, that means you don't want to go to where the ball is now -- you want to go to where the ball will be. Data aggregators like HSReplay and ViciousSyndicate can give very detailed information about where "the ball" is now, because that's objective; what they don't offer is subjective analysis of where "the ball" will be.
Thus a site like TempoStorm, which will never beat HSReplay or even Vicious Syndicate on data, tries to predict the future. In this case that means identifying not the winning decks of today, but the deck (or decks) that will best counter the winning decks of today.
The problem I see with TempoStorm's most recent Meta Snapshot isn't that it tries to make daring predictions, but it deviates from matchup data (without sufficient cause). They claim that "Control Warlock really only has one true enemy right now on ladder, and that's Quest Rogue," but that's not accurate; Control Warlock has an unfavorable matchup against Cubelock, and this problem is worse at high ranks than at low ranks. Yet inexplicably they list Control Warlock as favored against Cubelock, despite a rather standard decklist and no expectation -- or more importantly, explanation -- of a particularly winning piloting technique. Not to mention bad matchups they chart against Spiteful Druid, Control Priest, Taunt Druid and Tempo Mage.
In my recent Standard meta predictions thread (https://www.hearthpwn.com/forums/hearthstone-game-modes/standard-format/217154-i-have-modeled-58-archetypes-heres-the-future), I also tried to predict who the Paladin Killer would be, but my answer was Control Priest due to its favored matchups against both Cubelock and Even Paladin. If I was managing Tempo Storm, I probably would have listed it as the #2 or #3 deck for next week... behind Cubelock, because it's simply not reasonable to assume it will be dethroned any time soon.
But to get an empirically 42% matchup as a 60% matchup... well that's strangely incompetent. I almost wonder if they're deliberately giving bad information.
control warlock is tier 1 in legend, people saying tempostorm is wrong are probably rank 20
I am envoy from nowhere in nowhere. Nobody and nothing have sent me. And though it is impossible I exist. ©Trimutius
One does not simply walk into Mordor,
unless they want to be the best they can be.