Failure when you must succeed

By RogueJedi, in Game Masters

On 15/08/2017 at 9:45 PM, Donovan Morningfire said:

My own approach in these cases is to take a page from the 7th Sea 2nd edition RPG and FATE Core.

Once the players pick up the dice, they are going to "succeed" in that they complete the task and the story moves forward. However, if they are unable to generate an uncancelled successes on their check, they instead "fail forward," essentially succeeding at a cost, something that makes the scene interesting without totally derailing the plot, with the potential to use any advantages generated able to be used to offset some (but not all!) of that cost of success.

It's worked out pretty well, and some of players have come to eagerly anticipate the other shoe dropping, while others await with dread what could possibly have gone wrong, and in some cases coming up with ideas (before someone else tells them to shush) that are far more sinister (or amusing) than what I've come up with.

Keeps the story moving (important!), keeps the players on their toes (also important but also fun), and avoids constant re-rolls until they succeed (boring!)

Interesting to see different GMing approaches. To me, this would wind me up no end; I would prefer to have the possibility of failure. What you've described here - succeeding at a cost - would, for me, come about on a Success with Threat result.

To be honest, I find the idea that Failure doesn't move the story forward somewhat strange. Think of Indiana Jones , where Indy fails all the time, but these failures (combined, crucially, with the actions of the antagonists or the nature of the environment) continue to drive the story onwards. James Bond also fails a great deal, and almost every mystery story ever is characterised by a sequence of failures that push the story onwards; think of an episode of something like House , where the characters fail repeatedly (sometimes with Threat, sometimes with Advantage) but that failure doesn't stop the narrative...because there is still a pressing issue to deal with, normally one that is worsening and, if they don't keep trying, will end in an ultimate failure.

The issue that I think a lot of GMs have is that they consider "Failure" the same as "Not Success". Failure isn't just an absence - it is a thing in itself, with consequences (whether immediate or longer-term). It shouldn't stop the story and leave everyone wondering what to do next; it should have ramifications that force the PCs to make a decision. If there are no consequences for failure, then I generally wouldn't bother with a check; as you said, repeatedly trying to roll the same thing over and over again is very boring.

In some ways, I wonder if the Advantage/Threat element of this system actually distracts from this; GMs see Failure with a load of Advantage and so don't want to have something bad happen. But that's not the point, at least the way I see it - failure is still a bad thing, and should still have the same bad consequences for the characters that it always did. Advantage just means that the characters can salvage something from that situation.

Of course, this is all just my personal preferences, not a slight on anyone else's approach. :D

I can't remember where these concepts come from, so if someone else can cite the reference, I'd appreciate the reminder . . .

Okay, I'm a player (and I happen to be one of Happy Daze players) but there are two types of campaigns/adventures: Sandboxes & Railroads .

I think the OP's complaint has to do with a "Railroad" style adventure where the players progress sequentially from one scene to the next written scene with no apparent deviation considered or, seemingly, allowed. (I don't like these types of stories as a Role Player).

This is also one of the hazards of running "canned spam" modules that are set on rails, because a failure can be so damning. (Not being vulgar here, I'm using the classical definition of damnation; which means to impede progress).

I like Wafrog's comment about using elements of an adventure for inspiration, because the Star Wars RPG, in general lends itself better to sandbox style game play. I think you will be more successful if you dismantle the railway lines in the adventures and present opportunities for adventure.

It's a lot harder as a GM to supervise sandbox adventures, because the PC's can change direction a LOT.

For example:

HappyDaze set up a story where our characters were given the opportunity to help a beleaguered ships crew defend themselves from a violent attack by boarders (while the ship was in port, planet side, and the crew incapacitated and unable to fight).

The obvious course of action would be to set up defenses and fight off the attackers!

Instead we moved the battlefield and hid it.

This was a pretty dramatic shift to our story and the inevitable fight over the macguffin has turned into a treasure hunt with the PC's holding the map.

Yes, we've hit some pretty solid walls (and doors) but we've been able to regroup, dust ourselves off, and find another angle to approach the problem.

Another bit of advice I picked up in the past; Do not include an encounter/trap/problem unless YOU as the GM can clearly see two easy methods for overcoming the situation. Usually your players will come up with a third ingenious approach to solving the problem. If you're relying on a "canned spam" adventure and there IS a binary succeed fail point, rewrite it!

And lastly, I recall in RTOJ when R2-D2 was blasted by a Storm Trooper while slicing the back door, only to have Han Solo roll a Dispair on his check to have the second set of blast doors slam shut.

"Okay, George. We're still not attacking through the front door! Doesn't Chewie have an AT-ST with some Driver uniforms?"

Heck, the whole ground battle of Endor was a series of failures, complications, problems, and despair for the Rebels.

1 hour ago, Mark Caliber said:

Okay, I'm a player (and I happen to be one of Happy Daze players) but there are two types of campaigns/adventures: Sandboxes & Railroads .

I'd say those are the two extremes, and most people fall somewhere between them. I try to run what I'd call a "railroad sandbox". The players can do what they want, but the NPCs are also doing their own thing, which they will continue to do unless the PCs interact with them in some way. Some of the things the NPCs do will affect the PCs, and vice versa. I don't really care where or when the intersection points happen, but they will...

On 8/13/2017 at 3:31 PM, Vorzakk said:

The problem with this system, however; is that in addition to framing the success / failure axis as you describe, you also have to come up with what to do with any advantages or threats.

This actually makes it easier for me and my group, but I can see that I am fortunate in this. I think the best way to master this is to practice it, and to listen to some live play podcasts where it's done quite well. And of course most of us remember Fiddleback's Skill Monkey segments, which were essential to my understanding of the nature of those symbols. The responsibility for resolving these results lies with the entire group with the GM as final arbiter - the GM is most certainly not required to come up with everything -- that defies the very spirit of group storytelling. A GM can and should outright solicit a narration of the results of a roll, and oftentimes the pieces all fall into place as everyone adds a little bit to the narrative.

On 8/14/2017 at 2:58 AM, edwardavern said:

I guess it depends on your definition of "binary". If there is genuinely just one route "forward" and a single dice roll will determine success or failure, then sure, those can cause problems (although in certain circumstances they may be dramatically appropriate). However, absolute pass/fail checks are fine provided there are genuinely alternative options available should the PCs fail . In fact, I would advocate that these checks should exist - if the PCs cannot fail at something, there's no point in rolling: there's no inherent risk in the check.

Again, to take the locked door example: let's say the PCs fail their Skulduggery check, and the GM determines that they therefore cannot pick the lock at all (not a ruling I would generally make, but let's leave that to one side for the moment). That's a binary obstacle - you either pick the lock or you don't. But the story doesn't have to grind to a halt - the PCs now have a host of options available to them. They could try to break the door down (Athletics) or, if it's more substantial, attempt to blow it up (Mechanics). They could attempt to find another way in (Perception) or find someone in town (Streetwise) who has a key that they can steal (Skulduggery), buy (Negotiation), clone (Computers), acquire through convincing argument (Charm, Coercion, Deception), or acquire through violence.

By binary, I mean one or the other, no middle ground. I don't think we're talking about the same thing, but I do agree that your examples are more typical of the desired gaming experience by most groups. I do agree there's plenty of examples in cinema that defy this guideline, and of course Rule Zero precedes all rules.

"If Luke fails that Gunnery roll, everyone is dead." - fine climax to a story!

"If Luke fails to make a Brawn check, he can't get away from the dianoga and thus ask C3PO to ask R2D2 to shut down the trash compactors. Everyone is dead" - a pretty crappy end to a story. (Of course there are likely other ways out for the clever group, but our heroes would not have been so lucky. )

On 14.8.2017 at 2:49 PM, whafrog said:

Agreed. Except for the beginner box, I've never run a module as-is. They are idea mines, and that's about it. I'd go so far as to say if you try to run them as written, you'll create more headaches for yourself than if you heavily adapt it to your own campaign.

Yes. And that has been true for pretty much all published adventure modules I have run in the last 20 years, give or take. That is not to say that the writers are not good, just that each and every group has its own ways, and that a written module simply cannot be a fit for all of them. The more railroadey, the less I follow them, usually, as many of my players tend to go off the beaten path anyway. So, use them for ideas, plot points, NPCs, artwork and whatever works for you and your players, but never feel too bound to them.

First thing: always try to avoid do or die situations with only one solution. Maybe even try avoiding thinking about solutions in the first place, because players tend to come up with totally unexpected ideas anyway. And before saying "no" to anything, always think if you could say "yes, but ..." instead. I also learned a lot by playing a few sessions of Mouse Guard. Essentially, a failure does not mean running into a dead end or insurmountable obstacle, but just the story taking a new, maybe different path. Of course, that requires some creative thinking by the GM, but that is part of the job, I guess.

Always remember that you are telling a cooperative story. And in a (good) story, heroes do fail. Then they get up, dust themselves off and go on saving the Galaxy.

On 8/13/2017 at 11:31 PM, Vorzakk said:

I agree strongly with both of you; and good GM's have been doing this for decades. The problem with this system, however; is that in addition to framing the success / failure axis as you describe, you also have to come up with what to do with any advantages or threats. This is by no means insurmountable (and I enjoy the system overall in spite of it); but outside of combat, I find it to be an annoying and unnecessary complication.

There's always something that can go wrong!

  • 'lockpicking the critical door' test seems like it should be a check.
  • You litereally cannot fail the test, because if they guys cannot get into the bunker, the whole story falls apart
  • Clearly, the heroes must manage to breach said door (because plot ...ahem...the will of the force) so it's merely a matter of time taken
  • Hence, the sensible suggestion of "do you manage to crack the door before the patrol arrives?" or argubly better (unless the story demands covertness) "if the next patrol turns up before you crack the door, how many rounds of combat do have to hold them off for?".
  • To me, that immediately suggests a seperate axis for threat/advantage. Success/Failure is when the patrol arrives. Threat/advantage is what state they're in; tired (advantage)? surprised (triumph)? wary (threat)? Two patrols at once because you timed this for a shift changeover? (despair)
10 hours ago, themensch said:

"If Luke fails that Gunnery roll, everyone is dead." - fine climax to a story!

...And then we all died.

Don't do this one without your players' buy-in, though!

10 hours ago, themensch said:

"If Luke fails to make a Brawn check, he can't get away from the dianoga and thus ask C3PO to ask R2D2 to shut down the trash compactors. Everyone is dead" - a pretty crappy end to a story.

But depending on your perspective, the Dianoga let go because it knew the trash compacter was about to cycle as much as luke fighting it off - so luke gets throttled a bit, but doesn't die - the Dianoga realises it's got to skedaddle and drops its lunch.

Equally, that means there's probably another way out - goopy, slimey, and with a hungry dianoga at the other end, but infinitely better than ending up in 2d.

3 hours ago, Magnus Grendel said:

But depending on your perspective, the Dianoga let go because it knew the trash compacter was about to cycle as much as luke fighting it off - so luke gets throttled a bit, but doesn't die - the Dianoga realises it's got to skedaddle and drops its lunch.

Equally, that means there's probably another way out - goopy, slimey, and with a hungry dianoga at the other end, but infinitely better than ending up in 2d.

I knew as soon as I wrote this that we'd get into a semantics discussion about how Luke escaped the dianoga. So, I spent a little time thinking about what the dice said to portray this:

GM: "Luke, Make an Athletics check or this mysterious creature pulls you under the fetid water."

Luke rolls. "Okay, net failure but with two Advantage and three Threat. Can I spend the Advantages to say that I got a big breath of air and that something spooks the creature?"

GM: "Sure. You plunge under the water and struggle a bit but are able to come up for air. A strange rattling shakes the room and the creature is spooked and disengages. The walls start to close in as the trash compactor lives up to its name - there's my three Threat."

Of course a multitude of rolls could have brought us here, but that one seemed like the most likely circumstance.

Excellent point though - if the dianoga had another way out, so did our intrepid heroes. And I do agree with your point regarding having player buy-in at that final climax where it's victory or TPK. If everyone's having fun, there's no problem!

On 8/14/2017 at 2:58 AM, edwardavern said:

However, absolute pass/fail checks are fine provided there are genuinely alternative options available should the PCs fail . In fact, I would advocate that these checks should exist - if the PCs cannot fail at something, there's no point in rolling: there's no inherent risk in the check.

I'll admit you've given me food for thought - for me, this is second nature in adventure design, so I should have been more clear in my scope. If a single roll means everyone packs up their dice and goes home, that is poor form. I know this seems like a ridiculous guideline but I continue to be surprised at how often this sort of situation is written into even published modules by professional writers. Since I rarely, if ever , run other peoples' adventures without judicious hack & slash, it doesn't come to this at my table. I'll also admit that this is a two-way street and that a clever group should be able to steer around any such obstacles. I would suggest, though, that not all groups have 30 years of roleplaying experience and may not turn as easily to their own solutions.

I agree 100% in saying that without risk, there is no reward. There's little fun in sitting down to play a game where everyone knows they'll win, even Sorry has winners and losers.

On 14.8.2017 at 0:58 PM, edwardavern said:

However, absolute pass/fail checks are fine provided there are genuinely alternative options available should the PCs fail . In fact, I would advocate that these checks should exist - if the PCs cannot fail at something, there's no point in rolling: there's no inherent risk in the check.

I partially agree with you. IMO, Pass/fail checks are not bad, as long as they don't halt the game, or introduce trying until success is gained. I personally tend to allow just one roll for one task. If it fails, PCs need to change their approach, at least a bit. Personally I see no point of just rolling until you succeed. (I'm not saying those (roll until succeed vs. always succeed) are only possibilities.)

And as a small related sidetrack. The enlightenment regarding this came to me when I understood the "yes, and...", and I let go my need for logical solutions, and need for control in game. I also realized that the GM doesn't have to have alternative options thought out before hand, if GM is really ready to allow PCs try even some hairbrained ideas, and succeed with them. This was really hard for me. Negative side of this is that sometimes solutions to problems may end up being kind of stupid. But at least in our group, PCs like it.

On 8/17/2017 at 11:54 AM, edwardavern said:

Interesting to see different GMing approaches. To me, this would wind me up no end; I would prefer to have the possibility of failure. What you've described here - succeeding at a cost - would, for me, come about on a Success with Threat result.

To be honest, I find the idea that Failure doesn't move the story forward somewhat strange. Think of Indiana Jones , where Indy fails all the time, but these failures (combined, crucially, with the actions of the antagonists or the nature of the environment) continue to drive the story onwards. James Bond also fails a great deal, and almost every mystery story ever is characterised by a sequence of failures that push the story onwards; think of an episode of something like House , where the characters fail repeatedly (sometimes with Threat, sometimes with Advantage) but that failure doesn't stop the narrative...because there is still a pressing issue to deal with, normally one that is worsening and, if they don't keep trying, will end in an ultimate failure.

The issue that I think a lot of GMs have is that they consider "Failure" the same as "Not Success". Failure isn't just an absence - it is a thing in itself, with consequences (whether immediate or longer-term). It shouldn't stop the story and leave everyone wondering what to do next; it should have ramifications that force the PCs to make a decision. If there are no consequences for failure, then I generally wouldn't bother with a check; as you said, repeatedly trying to roll the same thing over and over again is very boring.

In some ways, I wonder if the Advantage/Threat element of this system actually distracts from this; GMs see Failure with a load of Advantage and so don't want to have something bad happen. But that's not the point, at least the way I see it - failure is still a bad thing, and should still have the same bad consequences for the characters that it always did. Advantage just means that the characters can salvage something from that situation.

Of course, this is all just my personal preferences, not a slight on anyone else's approach. :D

There's a difference between player success and moving the story forward. Success, from a player perspective, is both overcoming a goal and not having to pay for the effort. Moving the story forward simply means that a flubbed roll doesn't end the game.

Failures that kill the game/story are an issue with trad games like D&D or Call of Chthulhu . In CoC, if you fail that Library d100 roll, you don't get the critical piece of information that's needed to keep the story moving.

The Advantage/Threat element isn't a distraction if it fits within the narrative. It's a way to apply a cost (or a bonus!) to any action, even if the action is successful. Strain is a pretty useful tool in this regard as it's easy for a GM to justify applying Strain penalties and it's easy for players to recover Strain. If a GM and/or a player can make a logical argument that fits within the narrative, it's appropriate.

Failure isn't bad unless it ends a character's story in an unsatisfying way. It's important for the GM to act as a cheerleader for the PCs and to make sure that the players feel like they always have options. That's done by never saying "No*" to a player's idea. Rather, they should say "Yes" or "Roll your dice."

FFG Star Wars definitely asks more of GMs than other mainstream RPGs I've played. GMs aren't simply providing challenges that have a Success/Fail state. They're providing a loose narrative that players move around in and solidify with their choices. What the dice rolls mean depends 100% on what the players are doing. Despite the suggestions in the rule book and the GM screen, there are only a handful of times where the dice rolls are locked into a specific meaning (I'm thinking of combat). Rolling dice merely informs where the story goes; it doesn't determine if the players are "bad at gaming" or "unlucky."

*An exception can be made if said idea conflicts with the narrative rules of the established setting.

At weekend, I played a small indie game named Lasers and Feelings (game world we used was basically TNG Star Trek). How GM played it, was kind of eye opening. Every roll we made, moved story forward, regardless of it being success or failure. On failure, story moved to direction positive to PCs, and failure moved story to some other direction (and usually ignited our ship to fire). That style doesn't work directly as-is with SW, but at least in some situations it might work. I think it was quite close to failing forward style. I think I try to use "Failing Upwards: the JarJar Binks Story" method when planning future scenarios to our group.

9 hours ago, kkuja said:

That style doesn't work directly as-is with SW

I don't see why it couldn't - it just depends on how the group narrates the results.

While FFG SW game is an old school game, with plenty of dice rolling (as opposed to L&F and other games, IIRC, that tend to roll once per scene or so), you can look at the results of the whole encounter/scene instead of individual die rolls to determine if you fail or succeed. Not dificult to do.

Cheers,

Xavi

The number of dice rolls doesn't measure whether a game is Old or New School. Old School games - which are the evolutionary offshoot of war gaming - spend the majority of their rule sets on the resolution of combat. New(er) School games devote more rules to codifying other aspects of play, including complex resolutions of social encounters.

Any kind of encounter can be condensed or expanded, depending on the narrative emphasis the scene needs in order to be successful. Social encounters can be expanded by having players "attack" their opponents' Strain Thresholds by using Charm, Deception, and/or Negotiation skill rolls. Combat can be reduced to a single opposed all-or-nothing roll or even a single skill check if you want to simulate Han shooting Greedo.

Edited by Concise Locket
On ‎14‎/‎08‎/‎2017 at 0:39 PM, robus said:

It's hard to say as the results are prescripted. But there are any number of moments that could end badly for the protagonists. When they're trapped in the trash compactor for example. If R2 failed his slicing check then perhaps they would have become goo?

No because that's the "GM" failing to understand the system. Players must always fail forwards .

If I were beareded and a massive outsider whose distance from humanity would cause me to ruin my franchise will ill-thought-out prequels and rethinks of my original masterpiece-type GM, hypothetically speaking, I wouldn't make the computers check be to succeed in stopping things. I would assume that R2 already had access to the network; failure would be to say, route the commands in such a way as to make them unnoticed. i.e. can you do it without altering Imperials to the fact you're doing it and sending a squad to investigate (despair option).

Think of every time a film character in ANH, ESB or ROTJ objectively fails a roll. That is, their actions generated no successes at all. Examples could include:

  • Luke's computers check to lock the door behind him and Leia on the Death Star (also blasting bridge controls)
  • Han failing his sneak roll in ROTJ and alerting the scout troopers
  • Luke failing his negotiation checks with Jabba in ROTJ

Now, think of a "failure" in dice terms from the films and where it mattered to the pace of the story?

9 hours ago, Endersai said:

No because that's the "GM" failing to understand the system. Players must always fail forwards .

If I were beareded and a massive outsider whose distance from humanity would cause me to ruin my franchise will ill-thought-out prequels and rethinks of my original masterpiece-type GM, hypothetically speaking, I wouldn't make the computers check be to succeed in stopping things. I would assume that R2 already had access to the network; failure would be to say, route the commands in such a way as to make them unnoticed. i.e. can you do it without altering Imperials to the fact you're doing it and sending a squad to investigate (despair option).

Think of every time a film character in ANH, ESB or ROTJ objectively fails a roll. That is, their actions generated no successes at all. Examples could include:

  • Luke's computers check to lock the door behind him and Leia on the Death Star (also blasting bridge controls)
  • Han failing his sneak roll in ROTJ and alerting the scout troopers
  • Luke failing his negotiation checks with Jabba in ROTJ

Now, think of a "failure" in dice terms from the films and where it mattered to the pace of the story?

Failure doesn't always have to advance the story. Sometimes your players discover that their Ace was just one of the schmucks that died horribly attacking whatever super weapon the Empire/First Order had this week. It sounds harsh, but players can always make a new character.

6 hours ago, HappyDaze said:

Failure doesn't always have to advance the story. Sometimes your players discover that their Ace was just one of the schmucks that died horribly attacking whatever super weapon the Empire/First Order had this week. It sounds harsh, but players can always make a new character.

But that still advances the story, in my book. If everyone had to close their books and pack up their dice and go home, that's where I'd consider the pass/fail mechanism a failure.

Exactly. Dying should be heroic and memorable, and usually player choice. If anyone listened, I'm pretty sure Order 66 talked about PC death once... ( @DarthGM ).

If the player dies attacking the Imperial Superweapon Of The Week™, then it should be something they agreed to earlier and should serve the story at some point. The examples of what a player death should be, in the films, are when Dutch dies in ANH, or when Arvel Crynyd (I had to look that up, promise) crashes his A-Wing into the Executor 's bridge. Something happens as a result.

Saying "they die, it's harsh, let them re-roll" brings back d20-OGL game memories to me, and after the sweet nectar of narrative dice, that's a bitter taste... :P

If a character death isn't advancing the story in an interesting way it's almost definitely a badly handled death that probably shouldn't have happened in the first place.

On 9/19/2017 at 11:12 PM, Endersai said:

Exactly. Dying should be heroic and memorable, and usually player choice. If anyone listened, I'm pretty sure Order 66 talked about PC death once... ( @DarthGM ).

If the player dies attacking the Imperial Superweapon Of The Week™, then it should be something they agreed to earlier and should serve the story at some point. The examples of what a player death should be, in the films, are when Dutch dies in ANH, or when Arvel Crynyd (I had to look that up, promise) crashes his A-Wing into the Executor 's bridge. Something happens as a result.

Saying "they die, it's harsh, let them re-roll" brings back d20-OGL game memories to me, and after the sweet nectar of narrative dice, that's a bitter taste... :P

There's nothing about this system that takes random character death off of the table. It might take an absurdly high critical result, but it can still happen randomly.

Of course, my favorite take on this is, "Oh, you thought your character was the hero of this story and destined for greatness? I guess you should have rolled better. Make a new character and we'll see if this one is the one."

Edited by HappyDaze
Autocorrect sucks.

Once, in a starship combat gone wrong, the PCs' ship took a horrible beating. By all rights, they should have been blown to smithereens, end of story. But it was a ton more fun to narrate them floating in a listing hulk with atmosphere pouring into space through multiple holes and having them make a mad dash to save themselves and stabilize the ship. Then, the salvagers showed up....

As the PC's biggest fan, it was my job to offer another couple doors behind the one that closed and was welded shut behind them.

1 hour ago, HappyDaze said:

Of course, my favorite take on this is, "Oh, you thought your character was the hero of this story and destined for greatness? I guess you should have rolled better. Make a new character and we'll see if this one is the one."

This just feels like everything I disliked about d20 to me. And it's hugely disrespectful to the players, taking away their agency in part but diminishing their contributions in whole. I've GM'd and played this system, and would never appreciate or approve of such a cavalier approach to PC death in either role.

The players, if they're good, invest themselves into their character so that they're playing an avatar they want to play. They are connected to them. They enjoy them. They don't want to lose them. Mocking them with "oh you thought that? well no, try again, lulz" seems cruel, and perfectly suited to the D&D churn over story approach.

I can't agree with you here.

On 22/09/2017 at 5:58 AM, HappyDaze said:

There's nothing about this system that takes random character death off of the table. It might take an absurdly high critical result, but it can still happen randomly.

A lot of factors have to converge to get a critical roll of 141+. When things are tending towards that direction you've got a lot of control as a GM to course correct and provide alternatives to players if you don't feel like a death would be dramatically appropriate right now.

As far as I'm concerned, players should be able to tell a death is coming, and generally be able to take measures to mitigate the risk and avoid it. If a player comes out of a character death frustrated and disappointed, something's gone wrong. A death should feel deliberate, a real high point in the story that has big effects on the other surviving party members (if there are any), and a good end-note to that character's story arc.

I mean, obviously everyone enjoys different things and the whole cruel uncaring universe approach is what some people like, but it seems pretty unfitting to the tone of this universe and the rules of this game.

Edited by Tom Cruise