Animal repellent Puns

I need some help determining what statistical test would be appropriate for a research project I am designing testing the efficacy of 2 different commercially available repellents.

You have 10 animals in separate large enclosures. Inside their enclosure is a left and right hiding box. One of the boxes will contain repellent, one is a control. It is presumed that they would naturally go in and out of either or both boxes throughout the day, and would sleep in one of the boxes at night.

We will record with cameras and measure the time spent inside each box in minutes and record which box is slept in at night, if either. Each day/night cycle (say 23 hour period to allow one hour for setting up for the next day) is a replicate for each animal.

The order of treatments to be tested (repellent 1 or 2) and location of treatment (left or right) will be randomly determined for each animal (ex: Animal 1- t1 Left, t2 Right, t2Right, t1 Right, t2 Left and so on...). I don't know how many replicates will be necessary for each animal because without knowing what test to use, I haven't been able to do a power analysis. I think a biologically significant effect to detect would be if animals spent less than half as many nights in the repellent box and less than half as much time overall (those two things being analyzed separately, although unsure if overall time should include nights.)

What statistical test would be appropriate for comparing the number of nights spent in each box (or neither) and number of minutes in each box, and would it make sense to pool these results for each animal or to report them separately? From there, I should be able to do a power analysis to determine the necessary sample size.

I tried following this statistical test flow chart I was shown in class: https://www.scribbr.com/statistics/statistical-tests/

I believe the predictor variable (treatment vs control) is categorical. Then, I get confused. Is the outcome variable for number of nights spent in each box quantitative? The same for number of minutes in each box overall? Then, how many groups are being compared? Is it 10, each animal being a group? Then this would lead to ANOVA or MANOVA. But non-parametric tests like a sign test say they are used when there is a categorical predictor variable and a quantitative outcome variable.

Thanks in advance

So I spawn in and start running towards the stables and BAM a bear literally spawns 1/2 foot from my face and instantly kills me.

Now I know animals have been the worst now than they ever have been... but, it got me thinking... If Facepunch isn't going to fix their shitty AI animals. How about giving us a way to fix it ourselves?

The idea would be almost exactly how it's done with Barries. Use something already in-game (like Mushrooms) or add new items. Let us farm the seeds from this item and make Animal Repellent! Have it last 30 minutes just like the other teas and be done with it. Once used any animal in a X radius will just run away from you (unless you engage them).

The biggest issue I have since the animal update is not just them spawning near or on my face. But the biggest issue is that I could be trying to sneak up on someone or in the middle of PVP and BOOM pack of wolfs are attacking me from behind that weren't there a minute ago, or a bear spawns LITERALLY IN MY FACE while I'm engaged in PVP.

I really wish they had some stats tracking for AI deaths by animals (kinda like the shrine in Darksouls) because I've straight-up died more to these glitched-out crap AI's you call animals than I have in pvp...

Fix your game or give the players an option to fix it ourselves!

P.S.
Don't even try to say the AI isnt broken...
https://www.youtube.com/watch?v=57b2JoT_1zk