Before I get started, I want to say that I am not a statistician nor a mathematician. I could have very easily made a miscalculation or a faulty assumption. So take all of this with a drain of salt. If you see any ways to do these calculations better, put your corrections/calculations in the comments so we can all learn together! Cool? Cool.
So a common piece of advice that goes around on these forums is essentially "never concede, even if there is lethal on the board, because the opponent might miss lethal and you could end up winning." So I decided to ask myself "how much of an advantage are you giving yourself by not conceding?"
First, the assumptions: -Assume that you are an "okay" player with a 51% winrate -Assume you're playing a fast deck with an average game time of 5mins -Assume you play, on average, 1hr a day every day -Assume that conceding a game early saves you on average 15secs per game -Assume that the chance of your opponent missing lethal AND you then winning that game is 1%.
How many games do you play a year? 60 / 5 * 365 = 4380 games How many games do you win a year (w/ conceding)? 4380 * 0.51 = 2233.8 = 2234 games How many games do you lose per year (w/ conceding)? 4380 - 2234 = 2146 games How many extra games would you win by not conceding? 2146 * 0.01 = 21.46 = 21 games What would your winrate be if you never conceded? (2234 + 21) / 4380 = 0.5148 = 51.48%
So assuming your goal is to increase your winrate, not conceding will bump your winrate by about 0.5%, which isn't a lot but is still an increase. But what if you don't care about winrate but rather just the raw number of wins (i.e. if you were trying to hit legend)?
How much extra time do you gain by conceding early? (2146 * 15) / 60 = 536.5 = 536 mins How many extra games would you play by conceding early? 536 / 5 = 107.2 = 107 games How many of those extra games would you win? 107 * 0.51 = 54.57 = 55 games How many of those extra games would you lose? 107 * 0.49 = 52.43 = 52 games How many of those extra games that you would normally lose do you also win by not conceding? 52 * 0.01 = 0.52 = 1 game How many total extra wins would you gain? 55 + 1 - 52 = 4 games
You would only win an extra 4 games/year by conceding early, compared to our previous number of 21 games/year by not conceding. In other words, whether your goal is to increase your winrate or to get as many wins as possible (i.e. hit legend), it is in your best interest not to concede.
Now bare in mind if your goal is to play as many games as possible (win or lose), then yes conceding will let you play more games. And additionally, the amount of additional wins from not conceding is really really small (i.e. 21 extra wins PER YEAR). So at the end of the day, do whatever you want. But since I did these calculations for myself I thought they might be of interest to you.
Cheers, and best of luck on ladder!
Edit 1: Additinal thing to keep in mind is all of this is based off of the assumptions I made. If any of those assumptions are wrong the calculations would need to adjust accordingly. For example, I made a guess of 1% but I don't know how accurate that number is. Maybe it's way smaller (like 0.01%). Maybe it's way higher (like 20%). I think the main takeaway you should have is that ultimately the decision not to concede might give you ~slightly~ more wins but ultimately not a lot. So you should decide for yourself how much those few extra wins matter to you.
Edit 2: If you know of any data set that would give a more accurate estimate of the percentage of losses that would become wins if you didn't concede, please share it! It'd be super useful to get to the bottom of it (or if you're someone with research funding this could be a cool thing for you to study and make that data set).
Interesting math here, and it supports my own person credo to never concede unless you see lethal on board (and I mean obvious, elephant-in-the-room lethal, empty board 1-health opponent playing mage with 7 cards in their deck and a board of minions with 6 cards in hand lethal).
I've won a handful of games because my opponent missed an on-board lethal I thought was clear, and they go and trade, or similar situations. Once in a blue moon you get lucky and your opponent's lethal is stopped because of a DC (like one of my matches from last month... ugh!). Sometimes they order lul a combo, or trade sequence. Sometimes their RNG gets the better of them (I'm looking at you, Zul'jin shooting yourself in the face with every burn spell you played the whole game).
It's not a huge deal, since the higher up the ladder you get, the less likely you are to face mistakes like this, but unless you're power grinding for legend in the last 3 hours of the season, it's likely you're not saving time by conceding. It does help avoiding tilt from those who purposefully extend their turn for BM, but a quick break from Hearthstone can clear that up.
So yeah, pick your poison, but I'm of the mind to wait and see if my opponent shoots themselves in the foot or not.
Rollback Post to RevisionRollBack
Rage quitting: the best way to ensure your opponent knows they beat a giant baby.
Interesting analysis, but you've ignored something important -- the fun factor.
Yes, logically, winning the most games is the assumed goal, but sometimes it's better just to extract yourself from an unpleasant situation.
Twenty-one games per year is a tiny number compared to the number of games you might not play at all if you get so tilted and sick of Hearthstone that you have to take a long break.
Most important, though, I think this level of obsession with your overall record is unhealthy. If you don't feel good about finishing a game, you shouldn't do it. It's just a game, and the vast majority of us are not professionals.
Rollback Post to RevisionRollBack
"Why, you never expected justice from a company, did you? They have neither a soul to lose nor a body to kick." -- Lady Saba Holland
I laugh when Bomb warriors concede when I have lethal. Why don't they at least wait and see if they get lucky and I draw enough bombs to lose first? I have won a few games where my opponent had lethal, and instead 2-3 bombs go off and I win the match.
You never know what will happen, unless you are in a hurry, might as well just wait and see. You never know when someone might disconnect, or just miss lethal.
Interesting analysis, but you've ignored something important -- the fun factor.
Yes, logically, winning the most games is the assumed goal, but sometimes it's better just to extract yourself from an unpleasant situation.
Twenty-one games per year is a tiny number compared to the number of games you might not play at all if you get so tilted and sick of Hearthstone that you have to take a long break.
Most important, though, I think this level of obsession with your overall record is unhealthy. If you don't feel good about finishing a game, you shouldn't do it. It's just a game, and the vast majority of us are not professionals.
I don't think this is post is about the fun factor at all. Most players who are this concerned about data and minmaxing their playtime get their fun from efficiency. I'm not QUITE this sold on this approach, but I get what you're saying. Sometimes, it's just not worth waiting for your opponent to make you feel bad while they win in the hopes that something happens that allows you to take back a match you expected to lose.
Sometimes, it just feels bad to wait for that off chance that the other player is about to botch a win. It's a personality thing; some can wait forever to hope for a misplay, others can't bear it because they get tilted over that stuff. The data-first approach from the OP here doesn't exclude the fun factor, you just can't quantify such a subjective concept in math.
Rollback Post to RevisionRollBack
Rage quitting: the best way to ensure your opponent knows they beat a giant baby.
Interesting analysis, but you've ignored something important -- the fun factor.
Yes, logically, winning the most games is the assumed goal, but sometimes it's better just to extract yourself from an unpleasant situation.
Twenty-one games per year is a tiny number compared to the number of games you might not play at all if you get so tilted and sick of Hearthstone that you have to take a long break.
Most important, though, I think this level of obsession with your overall record is unhealthy. If you don't feel good about finishing a game, you shouldn't do it. It's just a game, and the vast majority of us are not professionals.
I agree with you. At the end of the day I think fun is THE most important factor. However, fun is an extremely subjective measure and what is fun for one player is not fun for another. I personally don't really care about min-maxing my winning, I care about making my own weird decks and doing alright with them. But if for you winning=fun then knowing the math behind conceding might be valuable to you. I did the analysis because I think it's fun to analyze things and because I wanted to see if the "never concede" wisdom has any weight to it. And all I can ultimately conclude is...maybe it does for someone?
Interesting analysis, but you've ignored something important -- the fun factor.
Yes, logically, winning the most games is the assumed goal, but sometimes it's better just to extract yourself from an unpleasant situation.
Twenty-one games per year is a tiny number compared to the number of games you might not play at all if you get so tilted and sick of Hearthstone that you have to take a long break.
Most important, though, I think this level of obsession with your overall record is unhealthy. If you don't feel good about finishing a game, you shouldn't do it. It's just a game, and the vast majority of us are not professionals.
I don't think this is post is about the fun factor at all. Most players who are this concerned about data and minmaxing their playtime get their fun from efficiency. I'm not QUITE this sold on this approach, but I get what you're saying. Sometimes, it's just not worth waiting for your opponent to make you feel bad while they win in the hopes that something happens that allows you to take back a match you expected to lose.
Sometimes, it just feels bad to wait for that off chance that the other player is about to botch a win. It's a personality thing; some can wait forever to hope for a misplay, others can't bear it because they get tilted over that stuff. The data-first approach from the OP here doesn't exclude the fun factor, you just can't quantify such a subjective concept in math.
From a scientific point of view it would be more credible if you had some data suggesting that you would win 1% of the games you do not concede. Because right now you just grab this probability out of thin air.
From a scientific point of view it would be more credible if you had some data suggesting that you would win 1% of the games you do not concede. Because right now you just grab this probability out of thin air.
That's totally fair criticism. I don't know of any existing data set that would let you know the actual % of winning games you don't concede, but if there was one that'd be extremely useful in making this more accurate.
Looks to me like that maths you present is based on the assumption that 1% of allgamelosses are due to having the opponent miss lethal and then you go on to win the game. Is that what you really mean? Or are you trying to say "given that the opponent has lethal on board, they will miss lethal and I will win 1% of the time" which would be very different.
Looks to me like that maths you present is based on the assumption that 1% of allgamelosses are due to having the opponent miss lethal and then you go on to win the game. Is that what you really mean? Or are you trying to say "given that the opponent has lethal on board, they will miss lethal and I will win 1% of the time" which would be very different.
I did my calculations based on the former, as I don't know of any data set I could use to distinguish in the latter. After all it'd be really hard to know what % of loses were missed lethal that "the loser" knew about, and then futher what % of those could have been won if they didn't concede.
I'd say 1% is way too high then. Games where I had lethal, missed it and then went on to lose are memorable because they happen so rarely. Games which are close are not the norm in this game.
Before I get started, I want to say that I am not a statistician nor a mathematician. I could have very easily made a miscalculation or a faulty assumption. So take all of this with a drain of salt. If you see any ways to do these calculations better, put your corrections/calculations in the comments so we can all learn together! Cool? Cool.
So a common piece of advice that goes around on these forums is essentially "never concede, even if there is lethal on the board, because the opponent might miss lethal and you could end up winning." So I decided to ask myself "how much of an advantage are you giving yourself by not conceding?"
First, the assumptions:
-Assume that you are an "okay" player with a 51% winrate
-Assume you're playing a fast deck with an average game time of 5mins
-Assume you play, on average, 1hr a day every day
-Assume that conceding a game early saves you on average 15secs per game
-Assume that the chance of your opponent missing lethal AND you then winning that game is 1%.
How many games do you play a year? 60 / 5 * 365 = 4380 games
How many games do you win a year (w/ conceding)? 4380 * 0.51 = 2233.8 = 2234 games
How many games do you lose per year (w/ conceding)? 4380 - 2234 = 2146 games
How many extra games would you win by not conceding? 2146 * 0.01 = 21.46 = 21 games
What would your winrate be if you never conceded? (2234 + 21) / 4380 = 0.5148 = 51.48%
So assuming your goal is to increase your winrate, not conceding will bump your winrate by about 0.5%, which isn't a lot but is still an increase. But what if you don't care about winrate but rather just the raw number of wins (i.e. if you were trying to hit legend)?
How much extra time do you gain by conceding early? (2146 * 15) / 60 = 536.5 = 536 mins
How many extra games would you play by conceding early? 536 / 5 = 107.2 = 107 games
How many of those extra games would you win? 107 * 0.51 = 54.57 = 55 games
How many of those extra games would you lose? 107 * 0.49 = 52.43 = 52 games
How many of those extra games that you would normally lose do you also win by not conceding? 52 * 0.01 = 0.52 = 1 game
How many total extra wins would you gain? 55 + 1 - 52 = 4 games
You would only win an extra 4 games/year by conceding early, compared to our previous number of 21 games/year by not conceding. In other words, whether your goal is to increase your winrate or to get as many wins as possible (i.e. hit legend), it is in your best interest not to concede.
Now bare in mind if your goal is to play as many games as possible (win or lose), then yes conceding will let you play more games. And additionally, the amount of additional wins from not conceding is really really small (i.e. 21 extra wins PER YEAR). So at the end of the day, do whatever you want. But since I did these calculations for myself I thought they might be of interest to you.
Cheers, and best of luck on ladder!
Edit 1: Additinal thing to keep in mind is all of this is based off of the assumptions I made. If any of those assumptions are wrong the calculations would need to adjust accordingly. For example, I made a guess of 1% but I don't know how accurate that number is. Maybe it's way smaller (like 0.01%). Maybe it's way higher (like 20%). I think the main takeaway you should have is that ultimately the decision not to concede might give you ~slightly~ more wins but ultimately not a lot. So you should decide for yourself how much those few extra wins matter to you.
Edit 2: If you know of any data set that would give a more accurate estimate of the percentage of losses that would become wins if you didn't concede, please share it! It'd be super useful to get to the bottom of it (or if you're someone with research funding this could be a cool thing for you to study and make that data set).
Interesting math here, and it supports my own person credo to never concede unless you see lethal on board (and I mean obvious, elephant-in-the-room lethal, empty board 1-health opponent playing mage with 7 cards in their deck and a board of minions with 6 cards in hand lethal).
I've won a handful of games because my opponent missed an on-board lethal I thought was clear, and they go and trade, or similar situations. Once in a blue moon you get lucky and your opponent's lethal is stopped because of a DC (like one of my matches from last month... ugh!). Sometimes they order lul a combo, or trade sequence. Sometimes their RNG gets the better of them (I'm looking at you, Zul'jin shooting yourself in the face with every burn spell you played the whole game).
It's not a huge deal, since the higher up the ladder you get, the less likely you are to face mistakes like this, but unless you're power grinding for legend in the last 3 hours of the season, it's likely you're not saving time by conceding. It does help avoiding tilt from those who purposefully extend their turn for BM, but a quick break from Hearthstone can clear that up.
So yeah, pick your poison, but I'm of the mind to wait and see if my opponent shoots themselves in the foot or not.
Rage quitting: the best way to ensure your opponent knows they beat a giant baby.
Most people concede because the game is frustrating as hell.
Interesting analysis, but you've ignored something important -- the fun factor.
Yes, logically, winning the most games is the assumed goal, but sometimes it's better just to extract yourself from an unpleasant situation.
Twenty-one games per year is a tiny number compared to the number of games you might not play at all if you get so tilted and sick of Hearthstone that you have to take a long break.
Most important, though, I think this level of obsession with your overall record is unhealthy. If you don't feel good about finishing a game, you shouldn't do it. It's just a game, and the vast majority of us are not professionals.
"Why, you never expected justice from a company, did you? They have neither a soul to lose nor a body to kick." -- Lady Saba Holland
I laugh when Bomb warriors concede when I have lethal. Why don't they at least wait and see if they get lucky and I draw enough bombs to lose first? I have won a few games where my opponent had lethal, and instead 2-3 bombs go off and I win the match.
You never know what will happen, unless you are in a hurry, might as well just wait and see. You never know when someone might disconnect, or just miss lethal.
#nerfbarnes
I don't think this is post is about the fun factor at all. Most players who are this concerned about data and minmaxing their playtime get their fun from efficiency. I'm not QUITE this sold on this approach, but I get what you're saying. Sometimes, it's just not worth waiting for your opponent to make you feel bad while they win in the hopes that something happens that allows you to take back a match you expected to lose.
Sometimes, it just feels bad to wait for that off chance that the other player is about to botch a win. It's a personality thing; some can wait forever to hope for a misplay, others can't bear it because they get tilted over that stuff. The data-first approach from the OP here doesn't exclude the fun factor, you just can't quantify such a subjective concept in math.
Rage quitting: the best way to ensure your opponent knows they beat a giant baby.
I agree with you. At the end of the day I think fun is THE most important factor. However, fun is an extremely subjective measure and what is fun for one player is not fun for another. I personally don't really care about min-maxing my winning, I care about making my own weird decks and doing alright with them. But if for you winning=fun then knowing the math behind conceding might be valuable to you. I did the analysis because I think it's fun to analyze things and because I wanted to see if the "never concede" wisdom has any weight to it. And all I can ultimately conclude is...maybe it does for someone?
Excellent point
From a scientific point of view it would be more credible if you had some data suggesting that you would win 1% of the games you do not concede. Because right now you just grab this probability out of thin air.
winning is fun, thanks OP
That's totally fair criticism. I don't know of any existing data set that would let you know the actual % of winning games you don't concede, but if there was one that'd be extremely useful in making this more accurate.
Looks to me like that maths you present is based on the assumption that 1% of all game losses are due to having the opponent miss lethal and then you go on to win the game. Is that what you really mean? Or are you trying to say "given that the opponent has lethal on board, they will miss lethal and I will win 1% of the time" which would be very different.
I did my calculations based on the former, as I don't know of any data set I could use to distinguish in the latter. After all it'd be really hard to know what % of loses were missed lethal that "the loser" knew about, and then futher what % of those could have been won if they didn't concede.
I'd say 1% is way too high then. Games where I had lethal, missed it and then went on to lose are memorable because they happen so rarely. Games which are close are not the norm in this game.
Even if someone misses lethal, it doesn't mean you're gonna win.