I don't know if it'll help, but ask him to look up step 5 ("choose kept dice") of making a check, since adding a kept die is done explicitly after step 5. Step 5 ends with removing all dropped dice. In other words, at the end of step 5 your dice pool only contains kept dice as the dropped dice are discarded with their results unused. You don't have any rolled dice left to look through to potentially add to your kept dice, because you only have kept dice at that point.
Edit: also, "set to" literally means the die got set, not rolled, to a specific result.

I've played more MC characters than GC. I'd hoped for a little more love.

Why is that questionable? You can remove Dying, but that doesn't prevent it from having happened. At the end of the round the Dying condition was still inflicted.

The +2 to TNs applies to any and all checks for movement. If an action involving moving doesn't have an associated check it doesn't get one due to entangling terrain, it stays check-free.

I don't think "affects your unarmed profiles" is all that vague. Because of that, I think this particular wording is a deliberate design choice and the Taoist ability is not meant to be useable with anything and everything involving unarmed attacks.
That said, that leaves us with FFG likely having shot themselves in the foot with the Water Fist wording. Which, if true (meaning Taoists should be able to use it with a weapon in a one-handed grip), is something they should have picked up on before the Shadowlands book was even finalized: if you're creating a new ability that you want to reference a set of pre-existing techniques, I'd expect you to doublecheck those after all. Whether they missed it or not, I doubt it will get addressed now. I suspect FFG is fine with letting GMs interpret/houserule as they see fit, rather than bother with ruleset maintenance for something that isn't so broken it makes the game or even just the school unplayable. It's not like a lot of effort has been poured into the faq/errata doc so far.

The point @UnitOmega makes is that Sharpened Ki is worded in a pretty legalistic way. Which is totally fine, because being carefully-consistent with rules wordings doesn't just serve to prevent ruleslawyering - it also makes it easier to write concise and, dare I say it, elegant rulesets (I don't think FFG really accomplished this with L5R). 
Using the "unarmed profiles" phrasing lets the designers create an ability that applies to a subset of kiho without having to list them (which would be bad, lists are inelegant design) and express a design intent for the ability not to apply willy-nilly to any and all kiho involving unarmed attacks.  

I don't think "affects your unarmed profiles" is all that vague. Because of that, I think this particular wording is a deliberate design choice and the Taoist ability is not meant to be useable with anything and everything involving unarmed attacks.
That said, that leaves us with FFG likely having shot themselves in the foot with the Water Fist wording. Which, if true (meaning Taoists should be able to use it with a weapon in a one-handed grip), is something they should have picked up on before the Shadowlands book was even finalized: if you're creating a new ability that you want to reference a set of pre-existing techniques, I'd expect you to doublecheck those after all. Whether they missed it or not, I doubt it will get addressed now. I suspect FFG is fine with letting GMs interpret/houserule as they see fit, rather than bother with ruleset maintenance for something that isn't so broken it makes the game or even just the school unplayable. It's not like a lot of effort has been poured into the faq/errata doc so far.

My sister did language studies at uni (same as our parents). There were, naturally, students from all over the country in her faculty - all with their own dialect from their home region. The ones from regions with the heaviest, most pronounced dialects were always the ones learning to speak according to the official norms the fastest and best; the ones with less noticeable dialects always took longer shaking those smaller deviations from the norm. 
Unicorns know they're still considered near-outsiders, and they're surely not above using their reputation for foreign and exotic customs to their advantage when it suits them. But they also know Rokugan is a highly formal society and that they have to work harder than anyone to conform. Being able to adjust and present themselves the very best way comes naturally to Unicorn diplomats. It's the dice rolls shaking out how they did, I know, but to me it's kind of fitting in its own way as well.

Well, yes and no. Unarmed profiles are basically weapons in all but name. In that sense the school ability is perfectly logical: kiho that affect a special set of "weapons" (unarmed profiles) can now affect a larger set (unarmed profiles and one-handers). The venn diagram concept kind of threw me because it doesn't seem to matter. That said, the Water Fist kiho now seems to suffer from the shoddy editing of the core book by not following the same template as the other elemental fist kiho - as written, arguably the kiho doesn't affect the unarmed profiles but the attacks made with them. I get that the Taoist Blade may not have been in development yet while the core book was being finalized, but that's kind of the point of editing consistency: it helps future proof the ruleset.

Well, yes and no. Unarmed profiles are basically weapons in all but name. In that sense the school ability is perfectly logical: kiho that affect a special set of "weapons" (unarmed profiles) can now affect a larger set (unarmed profiles and one-handers). The venn diagram concept kind of threw me because it doesn't seem to matter. That said, the Water Fist kiho now seems to suffer from the shoddy editing of the core book by not following the same template as the other elemental fist kiho - as written, arguably the kiho doesn't affect the unarmed profiles but the attacks made with them. I get that the Taoist Blade may not have been in development yet while the core book was being finalized, but that's kind of the point of editing consistency: it helps future proof the ruleset.

Well, yes and no. Unarmed profiles are basically weapons in all but name. In that sense the school ability is perfectly logical: kiho that affect a special set of "weapons" (unarmed profiles) can now affect a larger set (unarmed profiles and one-handers). The venn diagram concept kind of threw me because it doesn't seem to matter. That said, the Water Fist kiho now seems to suffer from the shoddy editing of the core book by not following the same template as the other elemental fist kiho - as written, arguably the kiho doesn't affect the unarmed profiles but the attacks made with them. I get that the Taoist Blade may not have been in development yet while the core book was being finalized, but that's kind of the point of editing consistency: it helps future proof the ruleset.

That's kind of my stance on most problematic mechanics. As long as it's for the sake of a fun and good game, we can put up with a little inconvenience.  In this particular case, if importuning as a non-shugenja (or pseudo shugenja) becomes problematic and ruins the fun just kick that out.

Shadowlands has more options so we'll likely see them in future supplements as well. I've always liked the heritage tables even though they've always been kinda random in their usefulness and power levels. My last game I let people pick, in the game I've just started running we rolled. Some results are problematic or powerful no matter what you do I found, but oh well so long as my players have fun who cares if it gives me the occasional headache justifying it in the lore/my game setting.

I'm not above picking a nit to prove a point and the editing of these books continues to be a disappointment to me, but I really can't bring myself to allow for the possibility that kicks and punches are not (meant to be) unarmed profiles.

I'm not above picking a nit to prove a point and the editing of these books continues to be a disappointment to me, but I really can't bring myself to allow for the possibility that kicks and punches are not (meant to be) unarmed profiles.

Yeah, this sounds like an abuse of shoddy semantics here... Punches and Kicks ARE unarmed profiles, whether the words "unarmed profile" appear in the description or not. It would also apply to Water Fist, too. All those Kihos are restricted to punch and kick (leaving out the only other unarmed profile: bite). There is no other way to make an attack with a punch/kick than to use an unarmed profile.

It gets kind of iffy moving resist checks to the end of the turn as well though - I mean, the outcome of that check can be beneficial to the attacker as well and should certainly be of interest to the attacker in the first place. I think the bigger issue is what opportunities can allow you to do and especially which opportunities can be created by certain types of checks. I think if resist checks get limited to success/failure outcomes, no opportunities (nor strife) that'd be a significant improvement already. Opportunities are great in general, but a number of specific instances are not good for the game.

How easily disadvantages are triggered (or advantages turned) is really dependent on the GM and on how the (dis)advantages get interpreted. I can see this working out really well in some groups, particularly ones that don't have a broad mix of characters, but arguably the GM needs to enable it to an extent. Then again, enabling the players is most of what a GM is supposed to do anyway in my opinion.

In this edition you never need ranks in something to be able to roll for it. School skill or not doesn't matter either. You just assemble a dice pool and roll, if you only have ring dice in the pool so be it.

In this edition you never need ranks in something to be able to roll for it. School skill or not doesn't matter either. You just assemble a dice pool and roll, if you only have ring dice in the pool so be it.

This is true. The players might feel this is something of a cheap cop-out or be disappointed with the results of the events and think their character isn't good enough, but they shouldn't worry (explain this if necessary!). If Kakita Toshimoko pronounces you worthy, not a soul in the empire will question this - in fact, you'll be the envy of most other young samurai for having had the opportunity to show your mettle like that. As for the character not being strong enough, given how the tests are set up you'd have passed if you hadn't gotten unlucky. Pray for a better roll of the dice next time. 

With the pregenerated characters odds are decent they'll pass, and regular characters should be a bit stronger than those. Did those players with only one point just have poor rolls all the time? Dice are random, but I'd expect at least 3 points for everyone without some really bad luck. These are standard characters, created using the normal rules? Have they been using void points?
Regardless, it's a bit tricky. I don't like taking the chance of failure off the table, the players need to know things can go wrong, but for total newbies it would suck to experience defeat so quickly. Remember to keep in mind they'll get XP before the 2nd day of the competition. It's not a big deal, but it can help with one or two events - make sure they know what's planned on day two (their characters should certainly know that). Spending that xp on skills that will come up the next day might help them get a few more points. I think if I were you I wouldn't change the contest to make it easier to pass, but it's your table and your players. Regardless of what you do, you can always allow players to create a new character for the next adventure if they're a bit disappointed.
As for the second question: it'd be really weird for the PCs to want to help someone else win, unless maybe it's someone from the same clan. Wanting everyone to succeed is one thing (and Hitoshi will pass, unless a PC prevents it), but the victory? This is an extremely prestigious event. It's really not in keeping with bushido to want to tamper with the outcome, and I can't think of any argument convincing enough to make Toshimoko consider not having it proceed normally (or as normally as possible, given what's happened).

L5R 5th is a bit less crunchy than Shadowrun but it's certainly in the same ballpark as D&D 3rd ed or Pathfinder (then again, I tend to think of most RPGs as low to mid-heavy rules systems - it's not about how many skills, classes or spells there are, it's about how complex the mechanics everything is built on are). Skill use in L5R 5th is definitely more complex than in Pathfinder, for instance.
The big difference is that a Pathfinder or something similar has a fairly strict set of rules: if you want to do something, how you do it is more or less explicitly explained in a rule and it's by and large always the same. L5R 5th on the other hand, in terms of core mechanics, is a bunch of guidelines and a non-exhaustive list of examples. Whatever you want to do, I can probably point you to a page (or, in some unfortunate cases, several pages throughout the entire book) that tells you what should happen. How it happens often depends a lot on interpretation and ad hoc decision making. How TNs are determined is completely vague. Approaches are literally the GM deciding which ring you use based on how you tell him you want to do something (and he'll also have to decide whether that changes the TN you'll roll against). There are long lists of possible uses for opportunities, but the book makes it clear you can really use them however you want as long as the GM is ok with it. Whether an advantage or disadvantage applies (or applies if you flip it) to a given situation is something that will vary from table to table (and you're encouraged to create your own). A whole lot of effects depend on interpreting the norms and social standards of the empire.
Some of that was certainly already the case in previous editions as well. And it isn't necessarily a bad way to do things either. I really wish the guidelines were a bit better in several cases though (those variable TNs for sure) and things like the advantages and disadvantages being so inconsistent and open-ended are a pain. It's definitely a system that needs a GM to make it work. That doesn't mean it's a light system, however.
 
Eh.
Skill use in D&D: I use this skill, I have these modifiers to the roll which always work uniformly so I can just add them up from my char sheet, I roll 1d20 plus or minus whatever, the GM checks for success or failure, done.
Skill use in L5R fifth: I pick the most applicable broad skill available (or the GM tells me which it'll be), I explain my approach, the GM tells me which ring corresponds to that, whether there are modifiers to the dice pool, and picks a TN based on both general difficulty and approach, I roll my dice pool, I check what I want to (and can) keep in terms of strife and opportunities, success/failure is determined and I probably get to use opportunities (regardless of success), which are often modified by a tech or school ability, done.
I may have missed a step here or there, but based on skill use I would definitely not call D&D the the rules-heavy system when comparing it to L5R.

L5R 5th is a bit less crunchy than Shadowrun but it's certainly in the same ballpark as D&D 3rd ed or Pathfinder (then again, I tend to think of most RPGs as low to mid-heavy rules systems - it's not about how many skills, classes or spells there are, it's about how complex the mechanics everything is built on are). Skill use in L5R 5th is definitely more complex than in Pathfinder, for instance.
The big difference is that a Pathfinder or something similar has a fairly strict set of rules: if you want to do something, how you do it is more or less explicitly explained in a rule and it's by and large always the same. L5R 5th on the other hand, in terms of core mechanics, is a bunch of guidelines and a non-exhaustive list of examples. Whatever you want to do, I can probably point you to a page (or, in some unfortunate cases, several pages throughout the entire book) that tells you what should happen. How it happens often depends a lot on interpretation and ad hoc decision making. How TNs are determined is completely vague. Approaches are literally the GM deciding which ring you use based on how you tell him you want to do something (and he'll also have to decide whether that changes the TN you'll roll against). There are long lists of possible uses for opportunities, but the book makes it clear you can really use them however you want as long as the GM is ok with it. Whether an advantage or disadvantage applies (or applies if you flip it) to a given situation is something that will vary from table to table (and you're encouraged to create your own). A whole lot of effects depend on interpreting the norms and social standards of the empire.
Some of that was certainly already the case in previous editions as well. And it isn't necessarily a bad way to do things either. I really wish the guidelines were a bit better in several cases though (those variable TNs for sure) and things like the advantages and disadvantages being so inconsistent and open-ended are a pain. It's definitely a system that needs a GM to make it work. That doesn't mean it's a light system, however.
 
Eh.
Skill use in D&D: I use this skill, I have these modifiers to the roll which always work uniformly so I can just add them up from my char sheet, I roll 1d20 plus or minus whatever, the GM checks for success or failure, done.
Skill use in L5R fifth: I pick the most applicable broad skill available (or the GM tells me which it'll be), I explain my approach, the GM tells me which ring corresponds to that, whether there are modifiers to the dice pool, and picks a TN based on both general difficulty and approach, I roll my dice pool, I check what I want to (and can) keep in terms of strife and opportunities, success/failure is determined and I probably get to use opportunities (regardless of success), which are often modified by a tech or school ability, done.
I may have missed a step here or there, but based on skill use I would definitely not call D&D the the rules-heavy system when comparing it to L5R.

L5R 5th is a bit less crunchy than Shadowrun but it's certainly in the same ballpark as D&D 3rd ed or Pathfinder (then again, I tend to think of most RPGs as low to mid-heavy rules systems - it's not about how many skills, classes or spells there are, it's about how complex the mechanics everything is built on are). Skill use in L5R 5th is definitely more complex than in Pathfinder, for instance.
The big difference is that a Pathfinder or something similar has a fairly strict set of rules: if you want to do something, how you do it is more or less explicitly explained in a rule and it's by and large always the same. L5R 5th on the other hand, in terms of core mechanics, is a bunch of guidelines and a non-exhaustive list of examples. Whatever you want to do, I can probably point you to a page (or, in some unfortunate cases, several pages throughout the entire book) that tells you what should happen. How it happens often depends a lot on interpretation and ad hoc decision making. How TNs are determined is completely vague. Approaches are literally the GM deciding which ring you use based on how you tell him you want to do something (and he'll also have to decide whether that changes the TN you'll roll against). There are long lists of possible uses for opportunities, but the book makes it clear you can really use them however you want as long as the GM is ok with it. Whether an advantage or disadvantage applies (or applies if you flip it) to a given situation is something that will vary from table to table (and you're encouraged to create your own). A whole lot of effects depend on interpreting the norms and social standards of the empire.
Some of that was certainly already the case in previous editions as well. And it isn't necessarily a bad way to do things either. I really wish the guidelines were a bit better in several cases though (those variable TNs for sure) and things like the advantages and disadvantages being so inconsistent and open-ended are a pain. It's definitely a system that needs a GM to make it work. That doesn't mean it's a light system, however.
 
Eh.
Skill use in D&D: I use this skill, I have these modifiers to the roll which always work uniformly so I can just add them up from my char sheet, I roll 1d20 plus or minus whatever, the GM checks for success or failure, done.
Skill use in L5R fifth: I pick the most applicable broad skill available (or the GM tells me which it'll be), I explain my approach, the GM tells me which ring corresponds to that, whether there are modifiers to the dice pool, and picks a TN based on both general difficulty and approach, I roll my dice pool, I check what I want to (and can) keep in terms of strife and opportunities, success/failure is determined and I probably get to use opportunities (regardless of success), which are often modified by a tech or school ability, done.
I may have missed a step here or there, but based on skill use I would definitely not call D&D the the rules-heavy system when comparing it to L5R.