LT odds and hoard breaker cheatsheet

Theghouse

With Wanda coming and many hoards breaking soon, I thought I'd share something I've been tinkering with.  I made the first basic version a while ago to help judge if I had enough pulls to break for BRB, and enjoyed the challenge of making it robust enough to handle huge amounts of pulls and other different factors.  I think most people will find it pretty straight forward, but I'll detail some of the features below.

Hopefully there are a few people out there who will find this useful or interesting!

Here is the link: https://docs.google.com/spreadsheets/d/16KgOCOA9P3-yGykmbxDDPSrIYWoz2OWygLO7l9g8WJE/edit?usp=drivesdk

You have to make a copy to be able to use it.

The top left box is how many pulls you want to take.  If you want to see the odds for taking 200 pulls, put in 200.  If you're not sure how many pulls you should be trying, you can fill out the target covers and characters (more on these below) and the sheet will give you the expected number of pulls as a good place to start.  I thought a good limit would be 5000 which is double the amount currently used as a target to 550 any given set of three.  Depending on what type of device you are using, be careful playing around with bigger numbers.  It takes a bit of processing power doing these big calculations for thousands of possibilities at a time.

The box below that is what you want odds for.  This was primarily created for Latest 5* and Special Legendary stores, but if you want a good laugh at the dilution in the game you can also calculate odds for getting classic 5* or getting 4* out of LTs.  If you're interested in seeing 4* odds out of other tokens, please see: https://images.app.goo.gl/56KycYbfVQVAX3sA8

Moving up and right, this box is the number of target covers you want from the pulls you are going to take.

The box below that is how many characters you are targeting in the current set.

The box next to that is how many unwanted colors.  So for example, if you wanted to pull for Thor covers but you already have his green at 5, you can put in 1 unwanted color and it will show odds for the remaining 2 colors only.

The two red shaded boxes are the counts for number of 5* in classics and number of 4*.  This has no affect on latest/special store odds but will affect if you are going to use the classic 5* or 4* calculator.  I think these numbers are accurate as of today but make sure you double check the current counts if you going to use those modes.

And then underneath that is a pulls calculator. Putting in your hoarded tokens and CP will spit out your total number of pulls.  If the sheet is set to latest/special, it will total your LTs and 25 CP/pull.  If you are looking at a special store, remember to take out your LTs and use CP only.  If the sheet is set to classic mode, it will automatically ignore LTs and only calculate at 20 CP/pull.

So for example, if you had no covers already of Kitty and Okoye, and there was a store with Kitty, Okoye, and Wasp, and you wanted to see what it would take to get Kitty and Okoye champed, you would put 26 covers (13 for each) and 2 target characters.  And you can see what it would do.

If you wanted to see how many pulls you would need to champ all three Latests but you already have one at 5/1/2 build, you would put 31 covers, 3 target characters, and 1 unwanted color.

If you wanted to see how many classic pulls it would take to get your 5/0/5 Okoye champed, set to classic, 3 covers, 1 target character, 2 unwanted colors.  And then hold on to something... (seriously, never EVER try to champ a 5* using classics)

Remember this is just odds and probabilities.  There's always a chance you end up in the bottom 0.15%.

If you have any issues, just post and I can try to help.

Akoni

You weren't kidding about the processing power. I used my laptop for it and it got so hot while processing that I had to put it in front of a fan to cool it off.

Theghouse

Akoni said:

You weren't kidding about the processing power. I used my laptop for it and it got so hot while processing that I had to put it in front of a fan to cool it off.

How many pulls did you go up to?  Lol

Akoni

I originally left it at 300 to see what the outcome would be, but decided to stop and lower it to 130ish. I haven't been hoarding for that long. I also spent a bunch of CP on one of the special stores with disastrous results.

KGB

You must have a very old laptop or your using Internet Explorer or your laptop runs hot.

I just did 1000 draws on my 6 year old gaming desktop in Firefox and it was done faster than I could scroll down. I did several other runs at 500, 400, 350, 300 and they all finished very fast (2-3 seconds to fill out the table) and the fan on my CPU never even spun up.

KGB

P.S. When I was totaling up my LTs + CP's I expected the CP column to divide my amount by 25 so that when I entered 3000 cp I would see 120 pulls. But instead it just took the number I entered and assumed that was the total I was drawing (so it said I had 3450 draws instead of 570 for 450 LTs + 3000 CP).

P.P.S. If you wanted to see how many classic pulls it would take to get your 5/0/5 Okoye champed, set to classic, 3 covers, 1 target character, 2 unwanted colors.  And then hold on to something... (seriously, never EVER try to champ a 5* using classics) - This only took 3120 pulls. It's a big number not not as big as your comment would lead you to believe. In actuality, if you set your BH to Okoye it would take at most 399 draws before you covered her via shards. It would be nice if your spreadsheet took into account shards in some way. Even if just saying how many sharded covers you'd expect to get for the draws made. So when you do 300 draws in latest it should say you get 2.25 sharded covers.

Theghouse

KGB said:

You must have a very old laptop or your using Internet Explorer or your laptop runs hot.

I just did 1000 draws on my 6 year old gaming desktop in Firefox and it was done faster than I could scroll down. I did several other runs at 500, 400, 350, 300 and they all finished very fast (2-3 seconds to fill out the table) and the fan on my CPU never even spun up.

KGB

P.S. When I was totaling up my LTs + CP's I expected the CP column to divide my amount by 25 so that when I entered 3000 cp I would see 120 pulls. But instead it just took the number I entered and assumed that was the total I was drawing (so it said I had 3450 draws instead of 570 for 450 LTs + 3000 CP).

P.P.S. If you wanted to see how many classic pulls it would take to get your 5/0/5 Okoye champed, set to classic, 3 covers, 1 target character, 2 unwanted colors.  And then hold on to something... (seriously, never EVER try to champ a 5* using classics) - This only took 3120 pulls. It's a big number not not as big as your comment would lead you to believe. In actuality, if you set your BH to Okoye it would take at most 399 draws before you covered her via shards. It would be nice if your spreadsheet took into account shards in some way. Even if just saying how many sharded covers you'd expect to get for the draws made. So when you do 300 draws in latest it should say you get 2.25 sharded covers.

For the CP and LTs calculator, the sheet ignores LTs by design when you switch to Classic because LTs can't pull classics.  Also of note is that it doesn't take your total and plug it in the top left for you automatically, the CP calculator bit is just there to add it up for you right underneath. If you want to run the odds on your total, you have to change the number of pulls manually in the top left.  I took the exact example numbers you gave me and it showed 570 pulls...  If you take another look at this let me know if something still seems off.

For shards, there is some info on shards earned for the number of pulls in the data listing above the bell curve, row 11 .  If it's not displaying make sure it hasn't hidden or resized any rows somehow.

As for the Okoye scenario, I thought it was absolutely bonkers that it takes 900 more pulls to get 3 Okoyes of one color out of classics (3120) than it does to get 339 covers to 550 an entire set of Latests/Specials (2260).  I'm surprised that's not more shocking to you!

Akoni

KGB said:

You must have a very old laptop or your using Internet Explorer or your laptop runs hot.

It's definitely not new, but not that old either. I use my laptop for work so it's probably already being pushed with everything else I'm doing at the moment.

Sekilicious

Also, laptop vs gaming computer...

Srheer0

Thank you for your efforts doing this.  

If anyone were just pulling Latest Legends and not hoarding, I found out from experience you get about 10-16 covers of each 5star between the time they enter Latest and the time they leave latest.  

Have now switched to the hoarding style. Looking forward to when Sw5ch and whoever the next 5star enters LL tokens.  

How often does the Latest Legends roster get changed? Is it every 8-12 weeks?

BlackBoltRocks

Srheer0 said:

How often does the Latest Legends roster get changed? Is it every 8-12 weeks?

The recent ones have been 12 weeks, as we've had a 4/5/4/5/4/5 schedule without deviation.

I keep track here: https://forums.d3go.com/discussion/69869/current-latest-5-in-tokens-and-future-4-additions/p1

akboyce

Looks cool. If you are looking for feedback you might want to change how the rounded odds are calculated or let it say things like 6 in 10. Reason I say that is for my current situation it says I have 68.65% chance to get 44 covers which it says is approximately 1 in 1. The next row is 62.73% which it says is approximately 1 in 2. I can swallow 62% being approximately 1 in 2 but its a bit of a stretch to say 69% is approximately 1 in 1.

KGB

Theghouse said:

For the CP and LTs calculator, the sheet ignores LTs by design when you switch to Classic because LTs can't pull classics.  Also of note is that it doesn't take your total and plug it in the top left for you automatically, the CP calculator bit is just there to add it up for you right underneath. If you want to run the odds on your total, you have to change the number of pulls manually in the top left.  I took the exact example numbers you gave me and it showed 570 pulls...  If you take another look at this let me know if something still seems off.

As for the Okoye scenario, I thought it was absolutely bonkers that it takes 900 more pulls to get 3 Okoyes of one color out of classics (3120) than it does to get 339 covers to 550 an entire set of Latests/Specials (2260).  I'm surprised that's not more shocking to you!

My bad. I looked again this morning and realized I must have had a moment of dyslexia late last night because I put the CP total in the LT column and the LT total in the CP column

Yes, it's a bit nutty that it takes 900+ more pulls to get 3 specific covers from classics than to cover all 3 latest to 500. But on the other hand no one would ever have to wait that long because shards would do it first. What is interesting is that by the time you made that many draws (3120) you'd get 10 covers for every classic 5* so you'd almost completely cover every classic 5 from scratch. With 23 more sharded covers + 4* feeder covers too from so many 4* covers you'd probably finish them all to 13 covers and have a complete set of 5* champs which is semi-impressive in it's own right.

KGB

Theghouse

akboyce said:

Looks cool. If you are looking for feedback you might want to change how the rounded odds are calculated or let it say things like 6 in 10. Reason I say that is for my current situation it says I have 68.65% chance to get 44 covers which it says is approximately 1 in 1. The next row is 62.73% which it says is approximately 1 in 2. I can swallow 62% being approximately 1 in 2 but its a bit of a stretch to say 69% is approximately 1 in 1.

Thanks.

I played around with more "precise" odds, and what I found is that it got very messy looking when the leading number wasn't consistent.  For example, it isn't immediately recognizable at first glance which of these is the highest and which is the lowest unless you do some quick mental math:

1 in 3
2 in 7
3 in 5
4 in 10

Those not familiar with probability or great at math would guess either the top or bottom, but 2 in 7 (28.6%) is the lowest and 3 in 5 (60%) is the highest.

I'll play around with making anything over 10 percent "x in 10" as opposed to "1 in x" and see how that flows visually.

Theghouse

Srheer0 said:

Thank you for your efforts doing this.  

If anyone were just pulling Latest Legends and not hoarding, I found out from experience you get about 10-16 covers of each 5star between the time they enter Latest and the time they leave latest.  

Have now switched to the hoarding style. Looking forward to when Sw5ch and whoever the next 5star enters LL tokens.  

How often does the Latest Legends roster get changed? Is it every 8-12 weeks?

Lately the release schedule has been a new 5* every four weeks.  That means the Latest group changes every four weeks, with the new character coming in and the oldest character rotating out to classics.  This also means each character spends 12 weeks total in Latests.

Sekilicious

I think I figured my question out using the doc. I need to average ~1/day to make sure the latest are champed when pulling as you go. I think. Or I got confused. Either way.

akboyce

Theghouse said:

akboyce said:

Looks cool. If you are looking for feedback you might want to change how the rounded odds are calculated or let it say things like 6 in 10. Reason I say that is for my current situation it says I have 68.65% chance to get 44 covers which it says is approximately 1 in 1. The next row is 62.73% which it says is approximately 1 in 2. I can swallow 62% being approximately 1 in 2 but its a bit of a stretch to say 69% is approximately 1 in 1.

Thanks.

I played around with more "precise" odds, and what I found is that it got very messy looking when the leading number wasn't consistent.  For example, it isn't immediately recognizable at first glance which of these is the highest and which is the lowest unless you do some quick mental math:

1 in 3
2 in 7
3 in 5
4 in 10

Those not familiar with probability or great at math would guess either the top or bottom, but 2 in 7 (28.6%) is the lowest and 3 in 5 (60%) is the highest.

I'll play around with making anything over 10 percent "x in 10" as opposed to "1 in x" and see how that flows visually.

Yea that makes sense. Something like:

100% 10 in 10
90% 9 in 10
80% 8 in 10
70% 7 in 10
60% 6 in 10
50% 1 in 2

The jump will be a bit jarring but it only happens at one point and the accuracy gains should make up for it.

XandorXerxes

Theghouse said:

akboyce said:

Looks cool. If you are looking for feedback you might want to change how the rounded odds are calculated or let it say things like 6 in 10. Reason I say that is for my current situation it says I have 68.65% chance to get 44 covers which it says is approximately 1 in 1. The next row is 62.73% which it says is approximately 1 in 2. I can swallow 62% being approximately 1 in 2 but its a bit of a stretch to say 69% is approximately 1 in 1.

Thanks.

I played around with more "precise" odds, and what I found is that it got very messy looking when the leading number wasn't consistent.  For example, it isn't immediately recognizable at first glance which of these is the highest and which is the lowest unless you do some quick mental math:

1 in 3
2 in 7
3 in 5
4 in 10

Those not familiar with probability or great at math would guess either the top or bottom, but 2 in 7 (28.6%) is the lowest and 3 in 5 (60%) is the highest.

I'll play around with making anything over 10 percent "x in 10" as opposed to "1 in x" and see how that flows visually.

You can convert them all to "1 in x" but that might take some spreadsheet-fu.

Using the above example:

1 in 3
1 in 3.5
1 in 1.67
1 in 2.5

I'd probably only do it for the first 15-20 lines or so too, any more would be superfluous since they vary well enough on their own.

Theghouse

XandorXerxes said:

Theghouse said:

akboyce said:

Looks cool. If you are looking for feedback you might want to change how the rounded odds are calculated or let it say things like 6 in 10. Reason I say that is for my current situation it says I have 68.65% chance to get 44 covers which it says is approximately 1 in 1. The next row is 62.73% which it says is approximately 1 in 2. I can swallow 62% being approximately 1 in 2 but its a bit of a stretch to say 69% is approximately 1 in 1.

Thanks.

I played around with more "precise" odds, and what I found is that it got very messy looking when the leading number wasn't consistent.  For example, it isn't immediately recognizable at first glance which of these is the highest and which is the lowest unless you do some quick mental math:

1 in 3
2 in 7
3 in 5
4 in 10

Those not familiar with probability or great at math would guess either the top or bottom, but 2 in 7 (28.6%) is the lowest and 3 in 5 (60%) is the highest.

I'll play around with making anything over 10 percent "x in 10" as opposed to "1 in x" and see how that flows visually.

You can convert them all to "1 in x" but that might take some spreadsheet-fu.

Using the above example:

1 in 3
1 in 3.5
1 in 1.67
1 in 2.5

I'd probably only do it for the first 15-20 lines or so too, any more would be superfluous since they vary well enough on their own.

They are current all set to 1 in X but with no decimals.  I am tinkering with a slightly different setup as discussed above.

Theghouse

Sekilicious said:

I think I figured my question out using the doc. I need to average ~1/day to make sure the latest are champed when pulling as you go. I think. Or I got confused. Either way.

I think I answered you on Reddit lol.  To give yourself an 85% to pull at least the minimum number of covers needed (without considering cover distribution which complicates the math much more), 3 pulls a day is the minimum is shoot for.

Sekilicious

Ugh. Work on 4* champs it is. Thanks. I did ask on both forums, yeah

McPosterson

much work.  just redeem when tokens arrive.  no math needed