Ok so fair warning, there are a pile of tabs and some aren't organized great. I'll give a little rundown.

https://docs.google.com/spreadsheets/d/1l7PIuG3QfE1G8na8fp5BBuWgzX-MW-29bLH4cWUQ4jg/edit#gid=1875850501

Tab 1: Attacks. Lists all attacks at lv1, makes it easy to see and compare by a few metrics. If you are out of coin all the time (which seems to happen to some people), avg cost for an attack, coin spent per point of war score (at lv25) can be useful, otherwise not really worth it. Interesting ones are Value and Value/Rarity. Both are formulas on a scale: If you are often out of war items, and have lots of energy, value should look like this: pts/(energy+(items1.5)) .....the 1.5 makes Items more valuable. If you run out of energy but have tons of war items, pts/((energy1.5)+items) energy should be more valuable. If you usually have just enough of one to use the other, then take out the 1.5 entirely, or reduce it. Adjust this value until it feels right for you.

Next the influence and value/rarity: rarity data is combined with checking what items are in each attack to give it a score based on how easy the items are to come by, overall, from war deliveries. Value/rarity takes the value and influence scores to work out a number that should give you a good idea how attacks really can compare to eachother.

Tab 2 and 3 are my attack levels, and a clubmate's.

Tab 4: WAR DATA - This is all of the raw data collected so far from war deliveries. At first, I had people submitting info to me, but since I had no guarantee that they sent me ALL of the deliveries, that data is sectioned to its own sample, and viewable in the overall data as well. Likewise, at a certain point I stopped taking data from the club and just used my own at lv25, so I could be sure to capture full information. All data before the war update is in a separate sample as my post-update data. Also from ~738 to 1100 I have some data in S-X columns that I used to get some stats about what items it wanted me to use. More on that later. Current data ends at 1400.

Sheet breakdown: A is a multiple choice selection of the war item. B is a sum of the sale value of all items requested (Early on this kinda sucked, but now I have named ranged for everything, so an entry is very readable, like =Corn+hats2+table2) Column C and D aren't particularly used now. They were a metric of how much factory/production time went into any particular request. This data wasn't useful. F column is player level. G column is the "date" - all deliveries done during a single war use the same date, to make it easy to group. I usually leave a blank line after a war ends. H Column: I type YES if I fulfill the delivery request, and blank if I skip it. More on this later, but I use this to run some stats, and get data like: How much coin did I spend in X war?

STATS columns: These are interesting - my data showed that you can have an effect on the cost of delivery requests, by skipping expensive ones, and fulfilling less expensive ones, ALL delivery requests tend to become less expensive, on average. To take advantage of this, the next 5 columns are used to help me decide instantly if I should fulfil the request.

I "Diff" - looks up the average cost of the war item being offered and compares to the cost of the delivery, limiting the result to either data that shares the level, then accounts for item rarity (if no level entered, compares it to the overall averages instead). A positive number means that it is lower than the average cost*(filters to only data newer than War Update). J "Value" - Yes/No to make it visible if the number is positive or not. K "SlipValDiff" - This column is similar to the previous, but filters to overall data. L "worth" - Also makes K more visible. N "Agreement" - if both the restrictive but weighted and less restrictive data agree with eachother, it is likely that their advice should be followed. Caveat - if it says you should not, but is a very low negative, often worth ignoring that advice. Comparing their outcomes with the H column should give you an idea what to look for.

Tab 5 - WAR STATS: Rows 4-14 account for items on each war item. B - Overall average cost of delivery. C - count of each time fully valid data was entered for each. C and D are the now-ignored Time. E - % of overall war deliveries of each item (later in this block of text will be the Filtered ones that I use). F and G are measures to determine how much rarity will affect H, the weighted overall average. I "Average" This is the weighted average of data filtered only to Level "25", and only from after the war update. J is the average cost of data from the point where I could guarantee all deliveries were being noted. This section is used as a medium compare between the small subset of post-update data (~200 items), "Just level 25 and clean data" (~587 items, including the 200), and "Overall data" (~1400 items). K is the count of "lv25 and clean", L is the % of war deliveries. M compares that subset of data to the overall numbers. Columns N-Q are roughly the same as K-M, except they only cover clean data from before the War Update.

Columns S to X, the good shit This section is all data from just my level, and just since the War Update. S is the average cost of deliveries my level since the update. T is the count of each delivery my level after the update. U for % of deliveries or rarity. V, W and X are all to compare to different data: Overall, all lv25, and all lv25 before the update.

Z to AB This section has stats on War Repairs. Average cost, count, % of war deliveries since I started getting clean data.
AD to AG Data from War Repairs, after the war update. Cost, count, average chance or rarity of each, and comparing that to previous data.

ALSO IMPORTANT You can change the filters in O and P 15 and 16, which are dates and levels. If you want to start keeping your own data, changing the Filter Level will update ANYTHING in the entire sheet that filters by level, so you can change it to see how different things would look for you. If you don't plan to enter your own raw data, don't bother changing it! I suggest leaving the filter dates. P15 is the last war before the update, O15 is the first war after the update. I usually change them only for checking specific wars or ranges/samples to see if there are any trends. Annnnd you can probably ignore the rest of that sheet.

TAB 6 "WarRepair" - pretty obvious, data on war repairs. Similar to War Data but with much less need to choose anything.

TAB 7 "Stock" - I have 3 feeder cities. This let me figure out what I had in stock so I could make enough of everything to not need to spend time trying to find someone who had it, or buy it from someone. Also nice to track where I store all my expansion items.

Tab 8 "War Summary" - I was going to start nice little blips of data summary here for each war, to be able to compare them.....I'm taking a break, so no entries.

Tab 9 "Maths" - Mostly rough calculations. M15-Q30 lets you put in hours and minutes remaining in war, along with current energy, to get N20 - How much energy is left in war and O20 - How much energy if I use 2 boosts in the last 2 hours without getting hit by vamp.

Tab 10 - "Item Data" is mostly raw data on each item in game, as well as some formulas to determine what to make if you are the most bored person in the world (colums k-L). Some useful items in the U-Y Columns: price of 1-5 of each item. Turns red if larger than 3200, which is more or less average War delivery request.

Last two tabs were plans to track more data but I never got into them.

Ok well....good luck with it. Let me know if you have any questions! there's a lot of stuff there.