Thank you for providing us with an example of hyperbole, which is exactly what your statement of "it happened in the last 4 of 4 multiplayer games blah blah..." is.
If you spent 40 resources on skill checks during your last 4 games you need to work on your strategic approach to the game.. or you know, don't lie to try to make a point.
Y'know, if the only way you can find to win an argument is to accuse your opposition of lying, you're already out the door, but whatever. I'll also note there's a helpful quote function here on these forums so you don't have to be bothered to type out "blah blah...", in case you haven't noticed it.
But it's not hyperbole if you are reporting what actually happened, and I did. You said "will literally never happen, ever", and it has. I'm sorry your argument was so poorly supported that you have to resort to such accords to advance it, but since you've already shown yourself the door, that does open the rest of us to have a productive conversation about the mechanics of the discussion without your acrobatics distracting people.
So let's get busy...
TripleA's belief that you have to negate a potential draw down to the potential -4 draw ("you cannot reliably pay the 3-4 resources to get you above the potential -4 or -5 modifiers on all your skill checks") shows a wrong application of looking at the odds (which is what led me to my Vegas comment, and I'm glad to see TripleA recognizes their vulnerability and will not risk livelihood there!). Whether you're looking at Skill cards, Skill Talents, or (my actual favorite) a mixture of both, overcommitting to a test wastes your resources -- hmm, can't use that word here, might get misunderstood, let's say game currency instead -- wastes your game currency (the game currency in Arkham Horror, what you can spend every turn, are cards, resources, and actions).
As in any gambling state, you're trying to get the most expected return on investment at minimum risk to yourself. While we don't know what token we're drawing from the Chaos Bag, we do know the contents of the bag, so we can figure out the odds on any particular draw (and with an actual game state surrounding it we can even include all of the iconography draws, such as whether the Skull will give us -0 or -2 based on number of ghouls sitting on our forehead).
The analyses I've seen of the standard difficulty bag for Night of the Zealot put the general penalty around a -1.3; if anyone here is a statistician, I'd love to see a thread where you deep dive into the Chaos Bag for fun data crawls to see what we find! Quoting Max Dog from my Survivors review (Max Dog, you did excellent work! Thank you again so much!):
"For example, if your modified skill is +1 greater than the difficulty of the test, your probability of success is 62.5%; increasing your skill advantage to +2 (say by committing a card) increases your chance of success to 81.25%. This shows us that In general, you want to be at skill advantage of +1 or +2. Increasing beyond that still improves our chances of success, but not as dramatically as going from 0 to +1 or +2. In general, we want to aim for +2 if we can without spending too many resources."
It's not cost-efficient in game currency to pump every Skill test you do to negate the potential -4 in the Standard Chaos Bag (unless the last turn is here and you're sitting on dozens of game currencies to spend, of course). You don't have to spend 2-3 resources with your Skill Talent on every pull to get a positive return on your investment; improving your odds from even to +1 nearly doubles your chance of success (37.5% to 62.5%); going to +2 above target number gives you nearly another 20% chance of success (to 81.25%). Spending 2 resources to go from 37.5% to 81.25% chance of success is a great return on investment. That, of course, is also how much you'll get from investment of a Skill card (what serendipity!). Getting up to +4 above target number only increases your odds to 93.75%; that's two more resources or one more (appropriate) Skill for a lot worse ROI.
No matter your strategy, raising your bonus to +1 at least above target number seems to be the main goal of your commitments or Skill Talents. Getting it to +2 gets as close to sinking the deal as we can with a reasonable spend of game currency, but I'm far more interested in how just a 1 resource spend in a Skill Talent can nearly double your success. The Skill cards will get you from +0 to +2 with one game currency, but then they're gone, and they can only be played on the appropriate tests, which seems to be a great balancing point vs. the 1 resource investment from a Skill Talent. And don't be fooling yourselves that the opportunity cost of having too many Skill cards in your hand isn't one you are paying; I've already seen games where Daisy and Roland never had to make a single Willpower test (except for a Very Special Willpower Test in Scenario 2, which some of you may know of); the Roland player was griping about the pair of Guts in his hand starting half-way through the game (also, my coffee was cold by that point, so I was doubly annoyed). Remember, unlike resources, cards are a game currency that have a limit to their pool (8), and I've also seen games where players had to discard cards due to hand size (one was an Evasion-heavy Wendy, and our Skids and Roland were doing such a great job killing threats that she had a hand full of nothing for anything).
So take a look at your game state the next time you play. How often does your deck have 1 resource to spend on an important pull that would push your Skill value to +1 over the target? If you're finding that is the case on a reliable basis, then a Skill Talent in that Skill will actually give your deck a +50% chance on that Skill, and would be a great investment (only Roland has access to all 4 Skill increases through Talents, but be that as it may). If you're finding your key tests are usually at 0 to -1 vs. target and you don't have a resource to spare, then a Skill card will give you a great burst chance of success, but weaken the deck through dead draws in other circumstances.
What's extra interesting is that, perhaps, Skill-card heavy decks might be making the best use of Skill Talents to back up the innate situational nature of their hands. If much of your deck costs nothing, your board state costs will be quite low, and so you're more likely to have extra resources hanging around to buck a key test up to +1 or +2. While TripleA seemed terrified of the 2-cost investment in a Skill Talent as a waste of resources, it seems to be a relatively equivalent opportunity cost vs. having situational Skill cards in hand and nothing to use them on, especially if most of your deck is free. Really, it's not like you have to play one or another; like many strategies, combining them will supply the best overall viability vs. a random adventure climate.
I also liked the discussion on how Skill cards can vary in usefulness based on number of players; more players = more chances to share Skill cards (which is awesomely interactive and fun to help your friends). But then you're going to have to keep the party together to get the most value of that tactic; is that worth going up against the clock for not splitting up and covering more ground?