Guides

@AvataSho @xianvar Innnnnteresting. :thinking: This box: [img]http://i.imgur.com/eVFVwTm.png[/img] ... was added before a lot of the other stuff. It was kind of experimental, and led to the introduction of more awesomeness, but is now a bit like an appendix. Kinda cool and useful, but not 100% necessary and can be removed without any big inconvenience. So I'm thinking, that maybe that whole box needs to be reconsidered, and maybe replaced with something MOAR awesome? Something *predictive* sounds daunting. We've got the "best day and time to sell" boxes, and a graph showing the way the prices shift through the day and through the week, so folks can see when the best time might be to sell. Even if there's a dom battle on and so it says the best day to sell is whatever today is and that might be Wednesday, people can still look at the graphs and see that Saturday will probably be better. But that's not very complete, I guess? Like, it leaves it up to the user to be savvy, and maybe it'd be COOLER if the spreadsheet could do that for us? I'm gonna have a think about this. In the mean time, if anyone wants to make a copy of the spreadsheet and come up with something cunning, please do go ahead and then share the results with me! My spreadsheet knowledge is by no means exhaustive. :) Edit: There's also the issue of accuracy. I don't want the spreadsheet to make a prediction based on flawed data (maybe because there's not been enough consistent form submissions in a week or two?) and have someone sell all their dragons when they could have waited and sold at a better time, for example.

@Cassolotl I think everything in that box is fine, but it's mostly the "Sell adults now" update I think is a bit misleading. Which xianvar was pointing out. The box sometimes says "Sell now" although a person with limited fodder and time would be best to sell their fodder only on Saturday for biggest payout. (A person more active with more access to fodder however is going to want to pay more attention and sell at peak price times constantly through the week.)

I think its mostly that it is emphasized but the Best Selling Time and Day aren't.
I wouldn't mind getting rid of that row (the Buy/Sell now row) and adding a color key for the last recorded prices boxes.
Yellow when prices are below the average since rollover. This is a good time to buy. Useful for snipers and suppliers. A good color that isn't discouraging that lets people know prices aren't increasing at the moment.
Green when prices are above the average since rollover. This is a good time to sell. Useful for those looking to make money now off fodder. A good color to let people know the prices are increasing.

As far as "predicting" goes. I think my usage of the word is a bit confusing. I don't mean something super monumental XD. I was really just suggesting that there be a graph for the current week, with a trendline that stretches till saturday thats based on the data since sunday.

It does run the risk of misleading information since obviously the current rate of increase/decrease in price is sometimes so dramatic (either when a push ends or another push is right after another, or even if a flight sort of "drops out" and prices never increase as projected.)

Edit: Actually, this could be done with a calculation. It takes the difference between the first sunday entry and the average of all entries up to this point. Divides that number by the amount of time since sunday rollover to give us a rate. Then we multiply it by how much time is left till saturday to get a rough prediction of how high fodder prices may get by saturday.
Of course if a push is competitive, prices will skyrocket till saturday's rollover, but this might be a good number to guess what prices will be like saturday morning.

@AvataSho [quote]I think everything in that box is fine, but it's mostly the "Sell adults now" update I think is a bit misleading. Which xianvar was pointing out. The box sometimes says "Sell now" although a person with limited fodder and time would be best to sell their fodder only on Saturday for biggest payout. (A person more active with more access to fodder however is going to want to pay more attention and sell at peak price times constantly through the week.) I think its mostly that it is emphasized but the Best Selling Time and Day aren't. I wouldn't mind getting rid of that row (the Buy/Sell now row) and adding a color key for the last recorded prices boxes. Yellow when prices are below the average since rollover. This is a good time to buy. Useful for snipers and suppliers. A good color that isn't discouraging that lets people know prices aren't increasing at the moment. Green when prices are above the average since rollover. This is a good time to sell. Useful for those looking to make money now off fodder. A good color to let people know the prices are increasing.[/quote] Okay, how does this look? [img]http://i.imgur.com/p631g5j.png[/img] I've removed the bits that were sort of put there as an experiment a while ago, and replaced it with a colour key and some tooltips. I've inserted some rows to space things out a bit more. I've made the colours (conditional formatting) in the latest dragon prices cells more sophisticated. I feel a bit better about this now! [quote]As far as "predicting" goes. I think my usage of the word is a bit confusing. I don't mean something super monumental XD. I was really just suggesting that there be a graph for the current week, with a trendline that stretches till saturday thats based on the data since sunday. It does run the risk of misleading information since obviously the current rate of increase/decrease in price is sometimes so dramatic (either when a push ends or another push is right after another, or even if a flight sort of "drops out" and prices never increase as projected.)[/quote] The more I think about it, the less a prediction makes sense. I can think of a bunch of ways I could do it, but I just feel like it's covered pretty well by the days-of-the-week average graph and the times-of-day average graph. I think a trendline would be less reliable than those graphs most of the time, especially during dom pushes or times when there have not been many form entries. It would all be an overly complicated way to say "sell on Saturdays, late in the day" - which I think the spreadsheet already kinda does! [quote]Edit: Actually, this could be done with a calculation. It takes the difference between the first sunday entry and the average of all entries up to this point. Divides that number by the amount of time since sunday rollover to give us a rate. Then we multiply it by how much time is left till saturday to get a rough prediction of how high fodder prices may get by saturday. Of course if a push is competitive, prices will skyrocket till saturday's rollover, but this might be a good number to guess what prices will be like saturday morning.[/quote] The hard part is making this work when there are very few form entries. So, say for example no one fills in the form on a Sunday. That's gonna mess things up. It all needs to work automatically without my input, and obviously a lot of things fall apart when there are very few form entries but *this* one would fall apart if there were no entries on a particular day of the week within the last week, which is a much more likely scenario! :S I'll give it a try though, and put the info in a slightly hidden location, and keep an eye on it. :) UPDATE: I have GIVEN UP because I find Google Sheets' attitude to datetime and queries too hard right now, and it keeps telling me the query completed with an empty output when there should be dozens output! However, I will keep thinking about it and if anyone knows how to make a query that says "give me everything in this range with a datetime that has happened since the datetime in *this* cell", please do let me know. :D

Hmm... would it work if you collected and saved data from several dom battles and then used past data to try and predict how the dragon prices will change?
You could have a graph that shows the average change in prices during similar times and whatnot?
Does that make sense? I have no clue how google sheets works so I don't know if it's possible or if what i'm saying even makes any logical sense
The problem that I find with my idea is that you'd need a way to either tell the sheet that it's a dom battle (like, through google form), or to have it figure out on its own (by looking at the dragon price increases).
Which is troublesome.

I don't know if my idea would work or be of any use but I might as well give the thread a friendly bump while i'm at it.

Oh and also a question about the form - what I've been doing so far when replying is actually looking for the average dragon price - like, if the very first dragon listed is cheaper than average and i know someone's already sniped it up - i just ignore it and grab the next one.
(i also groan becuase i missed a cheap dergon)
I'm going to keep doing that but I just wanted to... comment on it, I guess, and make sure I'm doing the right thing.

@sunbell

I think because things change so much even from dom battle to dom battle (like, depending on who's fighting who and stuff) it would be too hard. It needs to be able to work without my input, which means users would have to know that there was a dom battle going on - and if not, *I* would have to know that a dom battle was going on and tell the spreadsheet about it, which I just can't reliably do!

For the lowest dragon price, sometimes I miss off the cheapest one too if it's really obvious that it was priced deliberately low to sell instantly. Like if the top hatchling is 200 T and the next one is 4,000 T, I just put 4,000 T. :P So I think that's fine, really! At one point I considered asking people to enter the price of the dragon at the bottom of the first page of the search, but all that scrolling would do my head in... :D

@Cassolotl The new key works awesomely! I like the addition!

@AvataSho Yay, thanks! :)

I don't remember if i posted this already or not, but if not- this is still incredible and something I've wanted for the three years i have played on FR without really knowing how to have it :D thank you so much!

I have literally messaged my friends who don't even play anymore being like THERE IS A SPREADSHEET AND THERE IS GRAPHS, LOOK /heh

<3

@meowfish Wow that is very cool! :D Thank you for helping to spread the word, and I'm really chuffed that you like it! :)