Quote from: Lunamancer on October 13, 2022, 12:26:56 AMI appreciate the example, but I'm still at a loss here. ...I'm not sure why the players' "why" matters in any of this.

Let me try a different way of phrasing it: The "tactical" player is playing with an eye to what will be likeliest to "beat" the situation as a structured encounter resolved by the mechanics -- to maximize his odds by the rules. The "dramatic" player is playing with an eye to what will be likeliest to "beat" the situation as a described narrative resolved by GM interpretation -- to obtain success or failure by going outside the rules entirely, or by taking an action to completely change the situation the rules need to cover, and appealing straight to the GM's common sense, logic, or sense of drama to gain referee approval for that action.

Obviously there is lots of overlap between these approaches in practice, and they don't have to disagree on what they find to be the optimal solution to any one encounter. But they are different. And they can produce such disagreements. Being aware of that potential is something to take into consideration when designing games or scenarios.

In both cases, immediate and permanent character death as the result of a single failed decision point tends to be frustrating rather than entertaining, and so structuring the game to reduce the likelihood of this outcome is something I think worth doing. (You mentioned blackjack as an example of a game which still works despite regular, likely and immediate loss, but the catch is that in blackjack, you don't generally risk losing your pot for the entire evening on a single hand. And it takes seconds to reshuffle and deal a new hand, whereas it can take twenty to sixty minutes, sometimes much more, to create a new PC.)

QuoteMy experiences with Tomb of Horrors are right in line with what I typically see when running my homebrew stuff. Rules players are seldom effective. And when I first read Tomb of Horrors, I knew that's exactly how it would go, because in Tomb of Horrors it is to some extent done by design. It was meant to challenge characters with inflated stats. ...(but) it also has to be beatable by characters without inflated stats.

That is an interesting point which I didn't know, but doesn't that disqualify ToH as an example? All that shows is that typical PC rules-based strengths are irrelevant in an adventure which is deliberately designed to make them so, because it wants to subvert expectations created by typical play in that system. You can design one module, or module series, around that approach; you can't design an entire game system around it.

QuoteI am going to say something life has taught me is that for most problems and most situations, the significance of the things you don't know is greater than the significance of things you do know. Because that's how it goes when you live in a world of radical uncertainty.

I have no problem with introducing things not covered by the rules which one has to fall back on common sense to resolve. But introducing things that require non-rule-based thinking to resolve is not the same as making character capacities within the rules completely irrelevant to those characters' survival.

To consider your magical tuning fork, if one player's curious strike of the fork against a wall to see what happened was all it took to cause a cave-in so immediate and complete that it killed the entire party, I would call that counterproductively dangerous -- but I'd be willing to bet that wasn't the actual consequence in play, was it? Rather, I'd guess that any cave-in so caused would take the form of a series of saving throws which players could succeed at to escape, or to minimize damage such that those who did escape could save them. And if it did, then a player who correctly guessed there was a possibility of that outcome might decide whether to risk it or not based on his own knowledge of whether his saves gave him enough likelihood of surviving it -- and once again, knowledge of the rules becomes a critical part of the player decision process about what to risk.

Sooner or later, in any RPG, success comes down to a die roll against a target number, affected by modifiers (yes, sometimes you can find another option that doesn't require a roll, but not always). Deciding what to attempt based on quantifying the numbers that go into that is a strategy I think most people ultimately fall back on. Sometimes you can't know the actual numbers, and you have to guess their relative weight based on in-character assessment of the situation, as per your examples; I've never disagreed with that. But when you do know the numbers, or have the opportunity to learn them, I think people are likelier to decide by the numbers first -- what I've been calling "the tactical approach".

Conversely, what I've been calling "the dramatic approach" is the deliberate choice not to learn the numbers or think in their terms. This does not, as already acknowledged, mean that a "dramatic" approach will necessarily point to less effective options than the "tactical" one ... but it does open up room for that to happen sometimes.

QuoteYou should check out the prequel. It gets to a part where it just starts killing off all these Greek heroes, a lot of recognizable names. I was like, Holy shit!

Of course. But the Iliad is clearly the endgame scenario for a very long campaign for most of those heroes.  You wouldn't toss a bunch of 1st-level fighters, thieves and clerics into the front lines of a ten-year-old siege where the gods themselves are showing up and expect any of them to make it into Homer.

QuoteI ask players what it is they like most about playing RPGs and some of them answer, "It's all about a good story" ... These are normal gamers. ...And they experience stories in real time. Then I jump on line to read people insisting it's not a story, we've got some serious wires crossed somewhere.

Well, one of my criteria for calling a series of events a "story" (and I freely admit this is the perspective of an English major and occasionally published author, so not necessarily that of every gamer) is: Are the situations of the plot specifically tailored to the protagonist to dramatize his intended arc of character growth?

The Tomb of Horrors is the same dungeon for anybody who goes into it, barring excessive GM customization. If a player derives a meaningful character arc for his PC out of it, more power to him, but that's something he brought to the table, not something the dungeon as published was designed to provide him or facilitate for him.

So perhaps a better way to say it is, "Dungeon crawls are not stories until and unless the players choose to make them so."

Quote from: Stephen Tannhauser on October 13, 2022, 04:39:46 PM
Let me try a different way of phrasing it: The "tactical" player is playing with an eye to what will be likeliest to "beat" the situation as a structured encounter resolved by the mechanics -- to maximize his odds by the rules. The "dramatic" player is playing with an eye to what will be likeliest to "beat" the situation as a described narrative resolved by GM interpretation -- to obtain success or failure by going outside the rules entirely, or by taking an action to completely change the situation the rules need to cover, and appealing straight to the GM's common sense, logic, or sense of drama to gain referee approval for that action.

Obviously there is lots of overlap between these approaches in practice, and they don't have to disagree on what they find to be the optimal solution to any one encounter. But they are different. And they can produce such disagreements. Being aware of that potential is something to take into consideration when designing games or scenarios.

When I was out-tacticing the tactical player, the resolution was handled within the rules. It's not like it was a situation of a dramatic player winning at a drama-oriented game. However you try to define or distinguish what I was doing, the fact is I bested the tactical player at his own game. That doesn't suggest two different but equal approaches to playing the game. It suggests different echelons of play.

I never denied that playing badly can create problems. But bad play is something that can be corrected.

QuoteYou mentioned blackjack as an example of a game which still works despite regular, likely and immediate loss, but the catch is that in blackjack, you don't generally risk losing your pot for the entire evening on a single hand. And it takes seconds to reshuffle and deal a new hand, whereas it can take twenty to sixty minutes, sometimes much more, to create a new PC.

I assure you, I can afford to roll up more new characters than I can afford to lose hands of blackjack. It's a common but sad delusion shared by gamers that rolling up a new character is somehow costly. It's not. It's free. If taking an hour to do so is too much of a drag, that's a good reason not to play RPGs where character creation takes an hour.

QuoteThat is an interesting point which I didn't know, but doesn't that disqualify ToH as an example? All that shows is that typical PC rules-based strengths are irrelevant in an adventure which is deliberately designed to make them so, because it wants to subvert expectations created by typical play in that system. You can design one module, or module series, around that approach; you can't design an entire game system around it.

I think there's a difference between rules-based strengths are irrelevant and bloated stats won't save you. Playing characters on the lower end of the suggested range is definitely more challenging than playing one at the mid or high end. Stats matter in that sense. But you can't win without making wise choices. And bad stats won't kill you if you avoid the mistakes.

As to subverting expectations, I don't think that's a fair characterization of the module. Obviously to some degree traps need to subvert expectations. Tricking you is part of how they work. But no one would be able to reason their way through it if it was nothing but curveball after curveball. And that would undermine the point of it.

QuoteTo consider your magical tuning fork, if one player's curious strike of the fork against a wall to see what happened was all it took to cause a cave-in so immediate and complete that it killed the entire party, I would call that counterproductively dangerous -- but I'd be willing to bet that wasn't the actual consequence in play, was it?

Players didn't ding it except on the Bard. But indeed, one bad choice did have a small but not insignificant chance of a TPK, a few deaths at the very least. There was a good chance of it kicking off something of a mini-game of digging friends out while avoiding more falling rocks. And there was a slight chance of it amounting to nothing but a minor nuisance. A few areas in the dungeon were marked on my secret GM map as red zones, where dinging would have been especially disastrous and unforgiving. In one place, it would cause a collapse of the ceiling revealing a secret dungeon level that could easily be accessed by climbing up the rubble. The PCs never found that.

QuoteRather, I'd guess that any cave-in so caused would take the form of a series of saving throws which players could succeed at to escape, or to minimize damage such that those who did escape could save them. And if it did, then a player who correctly guessed there was a possibility of that outcome might decide whether to risk it or not based on his own knowledge of whether his saves gave him enough likelihood of surviving it -- and once again, knowledge of the rules becomes a critical part of the player decision process about what to risk.

The way it worked is if you dinged the tuning fork, there was a chance for either a total collapse (which would block the corridor) or a partial collapse. It's almost like the dungeon ceiling got to save for half. If it was a total collapse, everyone in the area would have to save or die. If successful, it still meant they were trapped. Trapped individuals could save themselves with a Bend Bars/Lift Gates roll (one chance only). If it was a partial collapse, a save indicated free and clear, a failure indicated trapped but alive. In all cases, of course, the rocks caused some damage, and all this is assuming you didn't die of falling rock damage.

So this player in your example would have been correct that a saving throw would save him from outright death in even the worst case scenario. But probably wouldn't anticipate that the severity hinged on the dungeon ceiling making an item save, and that in a worst-case scenario, if his Strength was too low to get a Bend Bars/Lift Gates chance, that there was no roll that would allow him to walk away without help.

QuoteSooner or later, in any RPG, success comes down to a die roll against a target number, affected by modifiers (yes, sometimes you can find another option that doesn't require a roll, but not always). Deciding what to attempt based on quantifying the numbers that go into that is a strategy I think most people ultimately fall back on. Sometimes you can't know the actual numbers, and you have to guess their relative weight based on in-character assessment of the situation, as per your examples; I've never disagreed with that. But when you do know the numbers, or have the opportunity to learn them, I think people are likelier to decide by the numbers first -- what I've been calling "the tactical approach".

Conversely, what I've been calling "the dramatic approach" is the deliberate choice not to learn the numbers or think in their terms. This does not, as already acknowledged, mean that a "dramatic" approach will necessarily point to less effective options than the "tactical" one ... but it does open up room for that to happen sometimes.

And I've never denied any of this. I just think it sounds better in theory than it works in practice. There's nothing assailable about your logic. No reason in principle you couldn't get some value out of partial information. In fact, I usually do it myself. But it didn't address the key issue I raised. That some people when faced with limited information over-emphasize the value of that information.

QuoteWell, one of my criteria for calling a series of events a "story" (and I freely admit this is the perspective of an English major and occasionally published author, so not necessarily that of every gamer) is: Are the situations of the plot specifically tailored to the protagonist to dramatize his intended arc of character growth?

The Tomb of Horrors is the same dungeon for anybody who goes into it, barring excessive GM customization. If a player derives a meaningful character arc for his PC out of it, more power to him, but that's something he brought to the table, not something the dungeon as published was designed to provide him or facilitate for him.

I'm a mathematician by education. But by vocation, I'm one of the top sales producers of one of the largest companies in the world. The secret to my sales success is to first understand the customer's perspective and what their motivation is, what problem they're trying to solve. And then to present my product to them in a way that demonstrates that it actually solves their problem.

The products I have to offer are the same for everyone. The needs of the customers are all unique. And my livelihood depends on the same product appealing to a unique motivation. And I'll be the first to admit. Not ever product is a fit for every person. But reaching an accommodation is a common enough thing that I have a job.

In sharing experiences about Tomb of Horrors, it comes through in spades. Sometimes someone says to me no way any party makes it through without losing a few members. And I'll say, "Really? I ran a group of newbs through it, and they did it without a problem." And they'll say, "There's no way they could have beaten Acererak!" And I say, "Beat him? You idiot. You don't have to fight him. He doesn't even do anything if you leave him alone." One of them hit me with, "Well, what if I'm playing a Paladin on a mission to destroy him."

Different motives. Different experiences. The module delivers.

QuoteSo perhaps a better way to say it is, "Dungeon crawls are not stories until and unless the players choose to make them so."

I forgot the main point of bringing up my sales background was that it's really, really important in my line of work to distinguish the difference between feature and benefit. Most people confuse the two. Gamers seem especially bad at it. Features are objective. Benefits are subjective. While the actual dungeon is a fixed, objective thing, the crawl is a subjective experience specific to the group. Why are you going into such a dangerous place? Motive. That's not a feature of the dungeon. And different motives generally imply different goals. Different win conditions. Different places within the dungeon you need to get to. What one group MUST accomplish by design of their own motives, another group has the option to bypass. The villain for one group might be the red herring for another.

I always thought one of the finest examples of a published dungeon ever, judged according to what I just laid out, was Keep on the Borderlands. By not having an overarching story, it stands aside allowing PCs to pursue their own goals. This makes the experience a lot more tailored to the PCs. And it's also got interesting bits for generating new story elements as you go. Like orcs try to capture PCs rather than kill. They let one go to fetch a ransom for the rest. So now you either have to raise the ransom money or plan a jailbreak.

In the TTRGP world it's regarded as perhaps one of the finest or at least most classic example of a sandbox. And I think sandbox is an apt term for it. But then the dictum gets handed down that sandbox and story are like oil and water. Why? There comes a point when it's fuck this shit. I experience what I experience and I know what I know.

Pluto got de-planeted. Pluto didn't change. What we know about Pluto didn't change. The definition of planet was arbitrarily changed. Word is, the vote that changed the definition was conducted improperly. It continues to be hotly contested. Years before it was taken, some people were already leaving Pluto off their models. Pluto didn't become any more or any less of a planet than it ever was. People just got pettier.

You can, in theory, reach zero health from maximum, depending on circumstances, but it's rare. Generally, you'd have to roll very poorly on defense, the enemy would need an excellent attack roll, and of course the lower your health total, the easier it will be. It's not that big of a deal though, since player characters don't permanently die, and there are plenty of ways to re-join the fight after being defeated.

Is there any meaningful difference between multiple consecutive rolls with no decisions in between and one roll with the same probability? I am not seeing the difference between roll to hit, roll for damage, save vs massive damage, and then save vs death compared to one roll with same probability of succeeding against some instant death effect. Unless there is a mechanic for modifying rolls in place I see no difference between one roll and many rolls to reach the same outcome without further player input.

Quote from: Wisithir on October 14, 2022, 07:35:28 PM
Is there any meaningful difference between multiple consecutive rolls with no decisions in between and one roll with the same probability? I am not seeing the difference between roll to hit, roll for damage, save vs massive damage, and then save vs death compared to one roll with same probability of succeeding against some instant death effect. Unless there is a mechanic for modifying rolls in place I see no difference between one roll and many rolls to reach the same outcome without further player input.

If there's no possibility of intervention or differentiation, then there is no difference.  Presumably, however, the sequence should be couched in terms that make the latter rolls have some kind of difference.  Bitten by poisonous spider, save or die immediately.  Not great, but there's at least the chance to recognize that, "Hey, we are fighting spiders.  Maybe we should run for it if we don't have any way to deal with poison ahead of time?"  Better, is bitten by poisonous spider, save or die soon thereafter.  Even better if the players were scouting and noticed the spiders before they had to run.  Of course, it takes some time to learn that kind of play.  Some people only learn it by strolling into a dark room whistling, when the spider drops off the ceiling and bites them, and usually in those cases dead now versus dead in 10 minutes is not going to make any material difference.

This is why, for example, when traps are sprung, unless the detection options are completely screwed up, I usually announce the results in two stages with a decision in between.  "There's an ominous, sharp click and the rattle of gears, quick, what do you do?"  Then depending on the reaction, there might be a different save or no save needed or no save even possible.  Depending on exactly what the character was doing that started those events, some options might seem to the player to be a better choice.  For me, that's a lot more fun than "Pit opens up beneath you impossibly fast, make Save X or take 4d6 damage to your 8 hit points." 

Quote from: Wisithir on October 14, 2022, 07:35:28 PM
Is there any meaningful difference between multiple consecutive rolls with no decisions in between and one roll with the same probability? I am not seeing the difference between roll to hit, roll for damage, save vs massive damage, and then save vs death compared to one roll with same probability of succeeding against some instant death effect. Unless there is a mechanic for modifying rolls in place I see no difference between one roll and many rolls to reach the same outcome without further player input.

I asked this question earlier up thread, with an important twist. But I think too many words, too many topics at once led to the OP misinterpreted the question.

Here's the twist. Multiple enemies each making their own attack before you can react is also an example of multiple consecutive rolls with no decisions in between.

If the goal is to give PCs a fair chance to do something to change the outcome, you can't solve it just by reigning in the effects of a single deadly attack.

If the goal is just to eliminate one hit kills, then that's fine. It just leaves me a little curious why it's strength in numbers is fine but strength in strength is not. What is it we're really trying to accomplish?

Quote from: Lunamancer on October 14, 2022, 10:18:32 PMMultiple enemies each making their own attack before you can react is also an example of multiple consecutive rolls with no decisions in between.

If the goal is just to eliminate one hit kills, then that's fine. It just leaves me a little curious why it's strength in numbers is fine but strength in strength is not.

For two reasons: 1) Multiple foes almost always allow far more opportunity to foresee their danger and evade them ahead of time; 2) As shown by Luke Crane's "Let It Ride" rule in Burning Wheel, the chance of a series of consecutive rolls all succeeding is far far lower than one roll at the same probability succeeding -- the "Let It Ride" rule is basically a statement that a player rolls once for any given goal and then goes with that result, rather than the GM making him repeat a series of rolls where he only has to fail once to fail at his overall goal.

Quote from: Stephen Tannhauser on October 15, 2022, 02:28:25 AM
For two reasons: 1) Multiple foes almost always allow far more opportunity to foresee their danger and evade them ahead of time;

That's vague, generic, and not really true. It's clearly easier to see a giant coming than it is 5 elves.

Quote2) As shown by Luke Crane's "Let It Ride" rule in Burning Wheel, the chance of a series of consecutive rolls all succeeding is far far lower than one roll at the same probability succeeding

This evades the context.  My comment was in response to a question that had a specific stipulation on the table. Which was, if I may rephrase it for clarity, what difference is there if it's a single roll, or multiple rolls whose cumulative probability is equal to the one roll. So for example, comparing something that takes one roll that needs to roll a 20 on d20 versus something that calls for 5 rolls, all of which must be successful, but only need a 10 or better on d20. The chance of either one happening is 5%.

Quotethe "Let It Ride" rule is basically a statement that a player rolls once for any given goal and then goes with that result, rather than the GM making him repeat a series of rolls where he only has to fail once to fail at his overall goal.

It's pretty basic math with no special GM insight. It's a generic statement, and I think GMs will always be better off using their own judgment as to when to call for a roll and when not to rather than following this rule.

I don't think there really is a difference between the one big roll and multiple smaller rolls, except maybe psychologically it might feel like the multiple smaller rolls are more fair, even though it's the same thing mathematically.

The answer should be that it takes multiple steps to reach that point. Multiple mistakes, errors of judgment, etc. Which, in a way, could be the same thing as the multiple die rolls. Let's do a thought experiment.

Suppose that there's a tomb, and the player enters, and there's multiple mistakes they need to make before they finally hit the final roll that kills them. Let's say there's a trap: the trap is hidden, the player needs to mess up several things, and the chance of them messing up each thing is the same as one of those aforementioned die rolls happening. Like they don't search for clues, they don't do anything when it activates, they fail the save to avoid the trap, they fail the save to avoid the save or die that comes from the poison, they die. Most of us would say that is fair.

Now suppose you take the probability of all those happening and compress it into 1 die and you make it so the PC has to make that save automatically the instant they just enter the tomb itself. Would that be fair? Just a 1% chance anyone who sets foot here dies on the spot.

Quote from: mAcular Chaotic on October 15, 2022, 07:11:31 PM
The answer should be that it takes multiple steps to reach that point. Multiple mistakes, errors of judgment, etc. Which, in a way, could be the same thing as the multiple die rolls. Let's do a thought experiment.

So this was more or less the point I was getting at. One die roll, many dice rolls, what really matters is whether or not there is the opportunity for an interposing decision. And if you want to ensure there is such an opportunity, nerfing one massive damage attack is not enough. There would also need to be some allowance for cases of multiple successive attacks.


Speaking for myself, I don't view multiple mistakes as being a must before a PC dies. I have no problem with there being a probability of death without further opportunity to make decisions when danger is clear and preset, up to and including automatic death in the case where the PC intentionally trips a clearly-baited death trap.

I also hold it's also kosher for one decision to lead to a chance for an instakill probability when the PC is already in a situation where there is a probability of death and the decision in question actually has the effect of reducing that probability.

Bear in mind death of the character due to old age is 100% in the cards. If the DM is calling for the periodic disease checks, as I do, a character of average CON has a life expectancy of under 35. Taking even very high risks going into the dungeon or other dangerous adventuring, if it comes with it the prospect of massive amounts of treasure of the kind that would enable sufficient "donation" to an NPC cleric who can perform a Cure Disease, is nonetheless trading up to better odds.

Every player character dies. Not every player character truly lives. I'd rather focus on making sure PCs have the opportunity to do something meaningful rather than merely the opportunity to escape something challenging or dangerous.

Wait, under 35? What do you mean? Most human age tables put characters dying at old age around 100 or so.

Quote from: mAcular Chaotic on October 15, 2022, 07:11:31 PM
Suppose that there's a tomb, and the player enters, and there's multiple mistakes they need to make before they finally hit the final roll that kills them. Let's say there's a trap: the trap is hidden, the player needs to mess up several things, and the chance of them messing up each thing is the same as one of those aforementioned die rolls happening. Like they don't search for clues, they don't do anything when it activates, they fail the save to avoid the trap, they fail the save to avoid the save or die that comes from the poison, they die. Most of us would say that is fair.

Now suppose you take the probability of all those happening and compress it into 1 die and you make it so the PC has to make that save automatically the instant they just enter the tomb itself. Would that be fair? Just a 1% chance anyone who sets foot here dies on the spot.

Does describing how or what clues one looks for still call for a test, or just consumes time and succeeds?
Is there a descriptive event when the trap activates and a chance to declare a reflexive actions like hit the deck, shield cover and direction, or leaping backwards or a straight save vs trap?
If other PCs avoided the trap can they intervene to affect the save vs death roll?
Is it mandatory to pass through the trap to complete the mission?
If no player actions could effect the rolls involved, I would far rather roll to save vs TPK and let the next batch of characters investigate how the first party perished on failure.

Quote from: mAcular Chaotic on October 15, 2022, 09:00:12 PM
Wait, under 35? What do you mean? Most human age tables put characters dying at old age around 100 or so.

That's maximum age. Life expectancy is a whole different thing. If you go through the disease table and calculate the probability that a character with 10 CON randomly contracts a terminal disease, and then take that small probability but compound it over 12 monthly checks per year, that will get you to somewhere in the early 30's. Which is historically accurate. But using these disease tables also captures an interesting and accurate nuance of life expectancy. The average number of years a 20 year old had left to live was the same as the average number of years a 40 year old had left to live. The idea being if you made it to 40 in the first place, you were probably healthier than average.

Characters with high CONs don't need to worry about it so much. The average length of time it will take to die of natural disease exceeds the max age of the human life span. Character's with very Low Con, as in 5 or less you can only be an Illusionist range, you've got less than 5 game years from the time of character creation.

Cautionary note for elves. If you don't have some means of dealing with disease, that 1200 year lifespan won't mean anything. The -1 to CON doesn't help you any either.

Quote from: Wisithir on October 14, 2022, 07:35:28 PM
Is there any meaningful difference between multiple consecutive rolls with no decisions in between and one roll with the same probability? I am not seeing the difference between roll to hit, roll for damage, save vs massive damage, and then save vs death compared to one roll with same probability of succeeding against some instant death effect. Unless there is a mechanic for modifying rolls in place I see no difference between one roll and many rolls to reach the same outcome without further player input.

If I have to roll four times, and failing any one roll results in death, my probability of survival is low.

If I have to roll four times, and I have to fail all of them to die, my probability of survival is high.

Basically, multiple rolls are an alternative method to alter probability distribution without having to introduce complicated resolution math. Each of these rolls, for example, could simply be "roll higher than 4 on a d6".

Obviously, we assume that the roll type and threshold are the same in each case; if all the rolls are different, we can say nothing about the probability generally, and the advantage of this method (simplicity, ease of resolution) is negated.

Quote from: allisonkaas on October 16, 2022, 10:56:20 AM

Quote from: Wisithir on October 14, 2022, 07:35:28 PM
Is there any meaningful difference between multiple consecutive rolls with no decisions in between and one roll with the same probability? I am not seeing the difference between roll to hit, roll for damage, save vs massive damage, and then save vs death compared to one roll with same probability of succeeding against some instant death effect. Unless there is a mechanic for modifying rolls in place I see no difference between one roll and many rolls to reach the same outcome without further player input.

If I have to roll four times, and failing any one roll results in death, my probability of survival is low.

If I have to roll four times, and I have to fail all of them to die, my probability of survival is high.

Basically, multiple rolls are an alternative method to alter probability distribution without having to introduce complicated resolution math. Each of these rolls, for example, could simply be "roll higher than 4 on a d6".

Obviously, we assume that the roll type and threshold are the same in each case; if all the rolls are different, we can say nothing about the probability generally, and the advantage of this method (simplicity, ease of resolution) is negated.

Yeah, but for those 4 rolls, you could just do a single d100 roll with the probability of failing all 4 rolls, or whatever your fail condition is. And that's much simple too. That's the question.