I suppose this my vary by difficulty, but let's talk about standard. When would you try an "optional" test (such as investigating for clues or location test to pass a story or even attempt to fight/evade) or just do something else (such as move to a location or draw a card or gain a resource). Would you attempt if you had a zero difference with the skill involved? +1? +2? When is it okay to risk a low difference check?

Edit: I feel The Circle Undone need a mention as haunted really makes "trying" not really a good option as failure has huge costs. You have to be in it to win it.

It depends on what the consequences are for failure (including token effects and on-failure effects like haunted, retaliate, alert or hurting another investigator). In the case of a token with an on-failure effect, I'd want to be able to succeed on that token, for instance.

Usually the first thing I do is check the scenario card for the special token effects. If there is something that has a bad consequence on failure, I want to keep that in mind. There are some really bad ones out there, that can spawn doom or monsters, or other nasty effects. So if you're in a scenario that has something like Cultist Token: -3, if you fail, draw the top treachery card in the encounter discard pile- then I really want to be up by +3 all the time.


Even if that isn't the case, it's generally a good idea to be able to pass most checks that you try, unless you are planning to get something by failing. Cards like Drawing Thin, Look What I Found, or Rabbit's Foot, -even Lucky! change the dynamic a lot. That's a post for another time. So lets assume for a second that you're not doing fail shenanigans, you're not doing any -cancel a token- effects, etc., which if I'm interpreting your post correctly was the intent. Failing a check still does have some cost, mainly the cost of an action. So the next thing to do is to consider what you can do with that action. Sometimes it's just a resource or a card. But maybe you have Dario out, then you could get 2 resources for an action if you havent used him this turn. Let's again, sort of assume that it isn't a special case.

Then I look ahead to next turn. Is there something in my hand that I would like to play, that I don't have enough money for, and won't have enough money for on my next turn? If I'm trying to play a 4 cost card, and currently have 3 resources, getting a resource isn't ultra important because I will get a resource during upkeep and be able to play my card anyway. But if you are sitting at 2 resources and looking to play a 4 cost card, then taking 1 resource looks very attractive, as you'll grab your resource this turn and get to 3, upkeep will put you to 4, and you can play your card next turn.

Just having an extra resource can be very beneficial if you are using Talents as well. If you are using talents, resources are on-demand stat boosts, which is very useful sometimes. Resources usually don't go away (although they can in some scenarios, or with Paranoia, etc), so they might very well come in handy later. And what definitely wont come in handy is a failed investigate action. So it should be clear that I value resources.... I also value cards. It's a card game, so you're going to need cards. There isn't much else to say. If your hand is running low, if you don't have a tool that you need.... drawing cards is going to be a must at some point. You'll need to keep in mind your weakness, whether it's a good idea to draw, especially on your last action, but still... drawing cards is essential to most characters and most builds. For however much I've dwelled on the importance of resources, cards are just as important. Card icons are also on-demand stat boosts, and can sometimes be far more effective than boosting with resources, if you're willing to give up the card. If you have an action to spare, and your hand isn't at 7+ cards, and you wont be crippled if you draw one of your weaknesses, then drawing a card is pretty much always a good idea.

So back to the core topic. Given that resources and cards are going to help my investigator in the future, when should I risk not getting them in order to attempt to gain some other advantage, like perhaps, a clue? Check your chaos bag. If you have a less than 50% chance to succeed, I wouldnt even think about it. For example, if it is.... I could get a clue if I draw a +1, 0, or Star.... don't even bother, just get the resource or card. Looking at the Circle Undone setup for standard, before adding any tokens for storyline or quest consequences etc, your chaos bag looks like; Star, +1, 0, 0, -1, -1, -2, -2, -3, -4, skull, skull, fail. For the sake of argument, lets just assume that star is auto-pass, even though it might not necessarily pass on all characters.

So we've got 13 tokens. -2 or -3 are common values for special tokens. For this first part, lets say skull is -2. If you are +1 over the difficulty, you have 6/13 chance to pass the check. 46%. I would not attempt this check. What if you were +2? your chances improve to 10/13. About 77%. Notice the huge difference here. At 77% chance to succeed, this is worth the opportunity cost of an action, IMO. I would attempt this check. Just for funzies, look what happens if you are +3: Your odds improve to 11/13, about 85%. An improvement for sure, but an improvement of about 8% when going from +2 to +3, compared to the improvement of 31% when going from 1 to 2.

Now lets do the same experiment with skull at -3. At +1 over the difficulty, your chance to pass is again 46%, I would not attempt the check here. At +2 over the difficulty, your odds are now 8/13, about 62% to succeed. This is above 50%, but it's still borderline. Some people might attempt the check here. A situational call, how badly do you need what you are going for? Is it worth the risk? Personally, I would probably not. Because... At +3 over the difficulty your odds improve to 11/13, 85%. For 1 more resource/icon/stat boost you improve your chances by 23%, a large boost. I would most likely take a resource or a card and attempt to secure the +3 over the difficulty attempt on my next try, or try to draw into something to improve my engine, rather than trying the 62% chance.

TLDR: be up enough that you can succeed on the Special tokens, your odds improve the most that way.

Hope that helps!

I was going to chime in when I had a second, but AWP pretty much has you covered. It is very unusual that you want to just press your luck and try a test that you have low odds of passing. There are always more efficient ways to spend the turn in most circumstances. Even if that is just gaining a resource or drawing a card (that may improve your chance of passing on your next attempt--say by committing that card).

You really want to be paying attention to your statistical success rate as noted above by knowing what is in the bag. Sometimes just committing that one extra pip/icon from the draw action could improve your chances by more than you would think.

Just want to chime in as well and hail @awp832's magnificent post - you've covered so many bases there and given a sense of the questions big and small that may come up. Speaking purely personally, as a separate perspective, rather than as a note of something additional, I think drawing a card / gaining a resource *should* give future tests a chance not to be non-boost affairs; that is to say, ideally when I play I want to be in a position where I can pick and choose the tests I take and how much I commit to them. One of the ways the game fights back against investigators is by trying to put us into a hole wherein we don't have the cards and/or resources we need to get out - and getting into that hole is often aided by a player overcommitting on a test they don't need to, or losing cards/resources to a tentacle token, or so on... If I'm playing and I feel the lip of the hole is near, that's normally a sign to prepare to swing momentum back in my favour by stockpiling cards/resources for a crucial test...

So yeah, just speculatively testing is normally a no-no, unless I am already +2/+3 or if *very particular* circumstance dictate it!

A lot of it comes down to how important the test is to pass. For instance, if you need only one clue to advance to the next stage, it may well be worth taking a lower-probability shot at it. Under most circumstances, though, going for clues aren't worth it unless you've got near a 50% chance, and 66% is better.