Graeme's page

If throwing sand in their eyes can blind them, wouldn't throwing a live spider or snake at them make them shaken? Unless they take their standard action to brush it off, they continue to be shaken....

I don't have anything that sets a precedent, but I do have a question now that I've seen the catching fire text from the environment section.

What DC is the reflex save for the spell? Is it a flat 15 (as listed in the catching fire environment text) or is it caster-based (as per pretty much every other spell-based DC)?

I thought it was caster-based, but I'm no longer sure. My planned pyromaniac gnomish bard may need a rethink, depending on the answer.

As a suggestion only, how about Weird Words costing 1 round of bardic performance per two words (round up) instead of one per word? So 1 round for 1-2 words, 2 rounds for 3-4 words, etc.

The only alteration to the arcane spell failure chance is the 1% per optional favored class bonus taken. Gaining medium armor proficiency doesn't alter the arcane spell failure chance of medium armor warn.

You can take 20 on picking locks, but not disabling traps, so trap DC will typically be the limiting factor. The DC for a magical trap is 25+spell level, and that's generally going to be the highest DC you'll encounter.

9th level spell gives a DC of 34, so a Disable Device of 29 is as high as you need (min of 1 means a fail by 4 or less, which you can retry) unless there are circumstance modifiers being applied, such as lighting penalties, or challenging circumstances (eg. in a sandstorm). If you can take 10 on the test, you only need a Disable Device of 24 (after all modifiers)

As an aside, we hit the same issue regarding temporary modifiers, but in the other direction, several weeks ago in a game session. In that one, my character took CON damage and the question was whether the reduction to hit points should be calculated based on what my modified CON score after damage, or based on multiples of -2 CON damage.

eg. CON 10 takes one point of CON damage. Based on RAW, there's no loss of hit points because it's a -1 penalty for every two points of damage, and so one point isn't enough to trigger it. Based on CON score, however, the -1 CON dropped the character down to the 8-9 CON range, which is a -1 penalty.

I asked the question here on the forum after that event and the consensus was that you work off the accumulate modifier (ie. -1 CON does nothing and -3 CON would be a net -1 CON bonus). That agrees with what Kazaan has been saying.

It's the flip side of the discussion above, with similar wording as to the effect of ability damage not actually affecting the score (ie. a temporary modifier), but having an impact on associated skills.

If you rule one way on temporary ability increases, then you need to rule the same way on ability damage which is effectively a temporary ability decrease.

Also, I feel that the 1.5 multiple for two handed weapon damage should apply to the bonus. So a +2 temporary strength increase is only going to add +1 damage, but a +4 temporary strength increase (eg. a barbarian's rage) would add +3 damage to two handed weapons (+2 bonus times 1.5).

As long as you meet the prerequisites, yes.

EDIT TO ADD:

I'm assuming that the ranger feats also meet the requirements for feats you get from other classes. For fighter bonus feats, that's okay, but that's not true for Wizard bonus feats which aren't compatible with the ranger combat style feats.

Don't forget the penalty in his next turn. He doesn't have a swift action (he used it as an immediate action) and he has to make a move action before he can attack or cast spells, so no full attack action, either. It's a class special ability and while useful, not particularly overwhelming.

The interpretation we've used is that you can specify where the 10' cubes are in the range, and the image can move within those cubes, but the location of the cubes themselves are fixed for the duration.

Shapeable apparently just means you must make all the 10ft cubes contiguous and that you can decide how they are arranged at the beginning of the spell.

Where does it say that they have to be contiguous? I would've ruled that they don't. Most of the time they will be, but I'm imagining a 'wack-a-mole' type scenario where you have an image popping up at random from holes in the ground over a scattered area - non-contiguous 10' cubes, with the image moving between them.

Depends on whether you call something that adds extra damage while you perform a bonus or penalty.

That was my first reaction, too, but then I looked more closely and Lingering Performance clearly states that the effects continue after you end your performance. Since Discordant Voice states it has the effect while you are performing, I took that to mean that it won't do the extra damage once you've finished, even if the effects of the bardic performance itself continue.

However, I agree it's arguable either way. My view is that bonus and penalty have a specific meaning in Pathfinder, and extra damage is neither.

I read it as Discordant Voice alters the bardic performance by adding bonus damage to all of them. All bardic performances I can find are (Sp) of (Su), so as long as the 30ft requirement is maintained, it should last as long as the other bonuses.

While most bardic performances are (Sp) or (Su), some of the archetypes have (Ex) bardic performances (eg. the Berserkergang performance of the Savage Skald archetype).

There's nothing wrong with developing custom tricks. The description for Handle Animal contains the following sentence:

Possible tricks (and their associated DCs) include, but are not necessarily limited to, the following

The Animal Archive added a set of additional tricks over and above the ones in the core rulebook. One of those is Hunt

Hunt (DC 20): This trick allows an animal to use its

natural stalking or foraging instincts to find food and
return it to the animal’s handler. An animal with this
trick may attempt Survival checks (or Wisdom checks if
the animal has no ranks in Survival) to provide food for
others or lead them to water and shelter (as the “get along
in the wild” use of the Survival skill). An animal with this
trick may use the aid another action to assist Survival
checks made by its handler for these purposes.

Developing a new trick, Scout, seems to be a variation on the Hunt trick and so I would suggest it's a reasonable DC 20 new trick.

I would say no it doesn't.

Discordant Voice is dependent on you using a bardic performance to create a spell-like or supernatural effect. Lingering Performance only comes into play after you cease performing.

So, the bonuses and penalties of the bardic performance continue for 2 rounds, but you're not using bardic performance for those 2 rounds so Discordant Voice doesn't come into play.

oO how is it you have a "dex based magus"?

Magus usually concentrate on one of DEX or STR for their combat abilities. If it's DEX, they usually take the Dervish Dancer feat so they can use their DEX bonus for attack and damage when using a scimitar. That's a dex based magus. It's one of the reasons there are so many scimitar-wielding maguses in the world....

Why does the ability to take no skill check penalty for being smaller rely on the chosen skill used?

Because it's more difficult for a small character to act in an intimidating way, but when taunting (ie. using bluff) it doesn't matter how big the opponent is.

I find this quite reasonable.

If an ability gives a bonus to demoralize, does it add to the roll, if Bluff is used to Demoralize?

If the ability bonus is explicitly to demoralize (and not intimidate), then that would apply to the bluff check used to demoralize.

so the .5 and the .5 that add up dont stack?

I don't understand why you think that. The favored bonus in question is 1/4 bonus to the arcane pool.

1st time: +0.25 arcane pool points (round down to 0)
2nd time: +0.5 arcane pool points (round down to 0)
3rd time: +0.75 arcane pool points (round down to 0)
4th time: +1.0 arcane pool points. At last, an extra arcane pool point!!!

Human Add +1 hit point or +1 skill rank to the ranger's animal companion. If the ranger ever replaces his companion, the new companion gains these bonus hit points or skill ranks.

Half-Orc Add +1 hit point to the ranger's animal companion. If the Half-orc ever replaces his animal companion, the new animal companion gains these bonus hit points.

can they not select these favored class bonus until level 4?

can they pick them if they opt not to have an animal companion?

1. Yes, I believe they can take them before level 4 - it just doesn't have any impact until they get an animal companion. If they choose skill poihts, they're still restricted to the maximum skill points in any skill being the HD of the companion.

2. Certainly. That would be for when they decide to retrain their hunter's bond to get a companion. Why they'd want to do that is a completely separate question....

Name Violation wrote:

Aasimar

Add +1/2 to the oracle’s level for the purpose of determining the effects of one revelation.

do you need to know the revelation, or can you get the bonus to a revelation (you don't yet know) early so when you pick it up later its better when you do know it?

I'm going to say the former. You do need to have the revelation to add the bonus to it. Just as you would need to have the requisite spell levels to take the Human bonus for Sorcerer. (you can't add a bonus third level spell known when you're only casting 2nds)

I disagree. The wording doesn't say you have to know the revelation. The Human Bonus for Sorcerer is explicit - it's a bonus spell of a level at least one lower than the highest level you can cast.

My understanding is that human barbarians quite often start to take the 1/3 bonus in Superstition from level 1, even though they can't have that rage power until level 2 at least. The bonus certainly makes no mention that they have to have that rage power before they can take the bonus.

All of the 1/X favored class bonuses means that you get that bonus after taking it for X levels. In the case of the Magus, every four levels of taking that bonus gives you one point to the arcane pool. You get nothing until you've taken it four times.

Yes.

I think the problem is the name - it's not sensing traps, it's reacting to a trap that's been set off.

Trap Sense (Ex): At 3rd level, a barbarian gains a +1 bonus on Reflex saves made to avoid traps and a +1 dodge bonus to AC against attacks made by traps. These bonuses increase by +1 every three barbarian levels thereafter (6th, 9th, 12th, 15th, and 18th level). Trap sense bonuses gained from multiple classes stack.

Rogues get trap sense because dealing with traps is their specialty, and that includes dealing with traps that have been accidentally set off.

Barbarians get trap sense because they're supposed to have quick reflexes to deal with situations that occur out in the wild. If you think of them as a quick-reacting class, then it makes sense.

I'm guessing that the context is for prerequisites for prestige classes or feats. For example, a Dragon Disciple needs to be able to cast spontaneous magic.

I don't believe there's an official ruling, but I would personally say they're prepared, not spontaneous.

I did think of that, but then you're wasting one part of the Arcane Trickster class - adding one to the caster level for each level.

I've just realised that with the SLA FAQ allowing SLAs to help qualify for prestige classes, it's possible to get into Arcane Trickster with a single level of a magic-using class. For example:

Half-elf (Drow Magic) - Rogue 3 and Wizard, Sorcerer or Magus 1

That meets the sneak attack requirement (+2d6), the Mage Hand requirement, and the 2nd level arcane requirement (darkness as a SLA) - all in the minimum number of levels to meet the skill requirements (disable devices 4 and escape artist 4)

Assuming a 15 point build, I was thinking:

STR 10 DEX 18 (15+2 racial) CON 12 INT 15 WIS 10 CHA 7

Traits: Reactionary (or any other trait giving +2 initiative), Magical Knack

Feats (to level 5)
1R) Toughness
2R) Weapon Finesse (via rogue talent)
3R) Dervish Dance
4<spellcaster>) +1 INT for stat bonus
5AT) Improved Initiative

What do you suggest as the most appropriate spellcaster to go with this idea? I was thinking Magus for the ability to cast while in armor, plus taking a second level later gives the option to combine sneak attack with spellstrike, but I'd be losing out on the arcana until I took a third level of Magus.

Would Wizard be a better option? The idea is that the magic is mainly supporting the ability to get into position to sneak attack. For example, Vanish followed by a sneak attack with Shocking Grasp.

What I'd consider doing is to point out to the trapspotter that they don't know what's on the other side (just like they wouldn't know what's on the other side of a door) so barging through without taking any precautions is potentially dangerous.

From there, yes, I'd give them a chance of spotting the trap. That is, they feel a breeze (for the flaps that lead to a cliff) or sense an acidic smell in the air (for the flaps that lead to an acid pit). However, since the "trap" is actually on the other side of the flap (which I presume are clearly visible), the perception check to spot them is modified as a consequence. Through a closed door is a +5 DC, so that's your upper limit. I would suggest +2 DC to sense something through the flaps. If they're running, I'd also consider the +5 DC for being distracted. Yes, trapspotter gives them an automatic attempt to spot the trap, but it only activates when they're within 10' of the trap. If they're running, that doesn't give them a lot of time to react to that knowledge - the +5 DC for distracted is a fair analogy of that situation.

I suggest you have a read of Brewer's Guide to Reach Clerics. The tactics explained in that guide are applicable to many other classes as well. Have a look at the corresponding discussion thread for more ideas.

Okay, got it. So an archivist can decipher a scroll or identify a magic item while on combat using a take 10. Overall, not something that will crop up very often.

That's the way we're playing it. What makes you think that raising a character's hitpoints to zero or above won't make them conscious? Remember, a character on zero hit points is staggered, not unconscious.

As the Spellcraft skill description says nothing for or against taking 10, I had assumed that a character could take 10 on spellcraft rolls out of combat, such as for identifying the properties of a magic item or deciphering a scroll.

But then I read the description for the Magic Lore class ability for the Archivist archetype of Bards

Magic Lore (Ex): At 2nd level, an archivist gains a bonus on Spellcraft checks to identify magic items or decipher scrolls equal to half his bard level and may take 10 on such checks. An archivist can use Disable Device to disarm magical traps as per a rogue's trapfinding ability and gains a +4 bonus on saves against magical traps, language-dependent effects, and symbols, glyphs, and magical writings of any kind. This ability replaces well-versed.

If it's a class ability for an Archivist to take 10 on Spellcraft rolls for identifying magic items and deciphering scrolls, does that mean that normally a character cannot take 10 on those rolls?

Graeme wrote:

Ssalarn wrote:

I worry about how that might impact balance over-all as well though. If Superstition is off, are you still gaining the benefits of Spellbreaker, Disruptive, Witchhunter, etc.? All of those abilities assume that you have Superstition, are in fact contiingent upon you having Superstition. If you're choosing not to activate the pre-req, can you use contingent abilities ?

If it's houseruled that you can choose to not use Superstition when you start rage, then clearly you can't use any rage abilities that depend on having Superstition. I can't see how you can rule otherwise.

The same way you can use sunder without having to use power attack.

I would view it more in the category of not being able to use a feat if you've lost one of the prerequisites. The rage powers in question have a prerequisite and if you don't have that prerequisite (in this case because you've chosen not to include it in your rage), you can't use those rage powers.

But I see your point - it's certainly a legitimate interpretation. Since we're talking about a houserule on allowing Superstition to not be included in a rage, it just extends to a houserule on whether it needs to be present in the rage for rage powers that have it as a prerequisite to be useable. Personally, I'd rule you can't use the other rage powers that have it as a prerequisite, but that's just me :)

I think you're right on the total penalty. The way you've described it, they're penalties for different things and would stack, even if someone tried to argue otherwise. It's possible that the designers assumed that anything of animal intelligence would be non-humanoid, but I don't think that is a valid assumption - consider the various types of apes.

Personally, I would rule that humaniod is anything that looks humanoid. By RAW, humanoid is a specific type or subtype, but I think there are many non-humanoid creatures that common sense would rule as "humanoid".

RAW: Humanoid type or subtype only. Arguably monstrous humanoid, too, but that's debateable

RAI: (My opinion only) Also anything that looks humanoid. Please note that most monstrous humanoids look humanoid, but others don't (eg. Minotaur vs Lamia). I'd also rule that humanoid looking monsters without intelligence (eg. zombies and skeletons) can't be feinted.

I worry about how that might impact balance over-all as well though. If Superstition is off, are you still gaining the benefits of Spellbreaker, Disruptive, Witchhunter, etc.? All of those abilities assume that you have Superstition, are in fact contiingent upon you having Superstition. If you're choosing not to activate the pre-req, can you use contingent abilities ?

If it's houseruled that you can choose to not use Superstition when you start rage, then clearly you can't use any rage abilities that depend on having Superstition. I can't see how you can rule otherwise.

So your barbarian has, 24/7, a significant morale bonus to all his saves vs. spells, SLA and (Su) abilities. When he rages, he makes the CHOICE to give up those benefits--just during this rage!--so that he can be healed.

My understanding is that rage abilities only apply while raging. That means a barbarian does NOT get the Superstition bonus 24/7 but only when raging. This thread is about whether he can elect to not take that bonus for a specific rage.

Personally, I think RAW is clear and he can't, but I don't see how it's broken to houserule otherwise, as long as it's clear that you have the ability for the entire rage or you don't have it for the entire rage - you can't change your mind during the rage.

Our group is almost universally using the sCoreForge spreadsheet

http://sourceforge.net/projects/scoreforge/

The only person who isn't is the person who doesn't have MS Office, as the spreadsheet doesn't work in OpenOffice or its descendants.

The spreadsheet isn't perfect (a few bugs), but overall it's quite comprehensive and an impressive piece of engineering. It even caters for a fair amount of customization, such as custom magic items!

Spiritual Weapon (scimitar) + Improved Critical (scimitar) + Vortex Spell revelation (Wind Mystery) = Pure Awesome

Sorry, but the Spiritual Weapon spell description includes the following sentence:

Your feats or combat actions do not affect the weapon.

No Improved Critical (scimitar) - sorry.

Xaratherus wrote:

Diego Rossi wrote:

Sun metal: "The target weapon ignites into flame that does not hurt the weapon or the wielder, but damages those hit by the weapon." The enemy is damaged by an effect, not the spell.

I fail to see that the two are separate. It's not like you're using Spark to ignite a fire; the spell itself is the flame on the blade, which is why it doesn't hurt the wielder.

The spell activate an effect: flames on the blade. It don't do any damage by itself.

It is the same as Magic weapon. It don't do any damage by itself. It simply add damage to something.

So, that logic would also mean that Dazing Spell won't work with Dragon's Breath because the breath weapon is an effect and not the spell itself. Similarly for Fire Breath (it gives the ability to breath fire, not an immediate breath of fire), and Touch of Slime (this coats your hand with slime, and nothing happens until you touch something)

Where does the spell end and the effect of the spell start? Taking Touch of Slime as an example, the spell itself does no damage. You have to make a melee touch attack with your hand (or weapon for a Magus with spellstrike), so your logic would be that you can't use Dazing Spell with this (or any other touch attack spell).

That certainly weakens its usefulness, but is it RAW or RAI?

How do you choose between which pieces of information to give? In particular, if a creature has DR/xxxx, when would you tell the players what the "xxxx" is? Similarly for any immunities. Tactically, that's probably the most useful information they can get, but should that be the first thing they get or do they need a higher roll before they get that over other things (like any special attacks)?

1. If a GM were to allow it, I would suggest it would take at least two actions - one action to open the potion (perform trick) and then one to drink it (not sure which, but probably perform again). The reason is the mount is going to have to use it's mouth to open the potion first before it can drink it. You may also need a fetch trick first to get the potion in the first place (and hope it gets the right one!)

2. I would say yes - probably again using the perform trick. However, since the command word is probably not a word in the language that the circlet gives to the animal, I'd suggest it wouldn't be a 100% chance of success each time.

All subject to GM ruling, of course.

I'm currently playing my first pathfinder character who happens to be an archery ranger. I made a mistake when I designed him and didn't give him a CON bonus but took a 12 INT instead. That's really hurting - he's the character that's been knocked unconscious more than anyone else and has twice been close to death from hit point damage. I second Krodjin's view to switch INT and CON - it really does make a difference.

Roc is subject to GM ruling, but since the bestiary does say that Ranger's can take them as companions, I think you should be allowed to have one.

Rocs taken as animal companions by druids or rangers are typically newly hatched birds—a baby roc is the size of a person and ready for flight and hunting within minutes of hatching. Unfortunately for druids seeking animal companions of legendary size, an animal companion roc is limited to Large size—still large enough for a Medium druid or ranger to use the flying beast as a mount.

I thought that may be what you were thinking of. It would make a great guardian for a druidic grove! But for an animal companion (which appears to be what the OP is about) it's not an option.

I would agree with Dabbler - as far as I can tell, this is legal. The wolf uses a standard action to 'cleave' bite, then gets a free trip, then moves on to the next adjacent target to bite...

This would be a nasty combination with Strong Jaw and then Cleave, Great Cleave, Combat Expertise, and Greater Trip. Assume you had four enemies adjacent; you could attack the first, trip him, get an AoO on him, then cleave to the next, trip him, get an AoO, and so on...

A powerful combination, but as Combat Expertise has a INT 13 prerequisite, I don't see it happening anytime soon. Wolves are cunning, but they start with a INT 2. Getting that up to INT 13 so it can take the Trip feat chain is a major undertaking....

I agree - I'm just pointing out that without knowledge of what they've done, you don't know how good a random number generator the die-roller is using.

I got the rand() function to produce good random results, but the way to do it was to not use the starting digits (which is what rand()*X is doing), but use the digits further down the string. That is, multiple by 1000 (say), then take the remaining fraction, and use that as your 0-1 random number instead of the original number. It produced random results with my little program filling the grid.

Not really feasible to roll that many dice, especially as I didn't have die of the appropriate size. I used a 24x80 grid and it didn't put a character in spaces where X+Y was an odd number. But that only became obvious when I let the program run to fill up that grid. I don't see dice doing that. Yes, I understand short term unusual distributions, but why would it NEVER generate an X,Y coordinate pair (that is, two consecutive calls of rand() with rand()*24 then rand()*80, truncating the fractions) where X+Y was an odd number? It looked nice, but it definitely had a pattern to it that was not completely randomized.

If you Wildshape into a form that can speak, you can do so.

That would include Elementals, or Parrots. Not most animals.

Not parrots. This is explicitly listed in the Wild Shape description (my bold)

Wild Shape (Su): At 4th level, a druid gains the ability to turn herself into any Small or Medium animal and back again once per day. Her options for new forms include all creatures with the animal type. This ability functions like the beast shape I spell, except as noted here. The effect lasts for 1 hour per druid level, or until she changes back. Changing form (to animal or back) is a standard action and doesn't provoke an attack of opportunity. The form chosen must be that of an animal with which the druid is familiar.

A druid loses her ability to speak while in animal form because she is limited to the sounds that a normal, untrained animal can make, but she can communicate normally with other animals of the same general grouping as her new form. (The normal sound a wild parrot makes is a squawk, so changing to this form does not permit speech.)

Umarian wrote:

Rynjin wrote:

The vast majority of dice will be just as random as a dice roller.

While I can agree with the first part of your post, this part is not accurate. It would depend upon the person that did the coding for the die roller. AS I said, the mathematical formula for the perception of randomness is extremely complex.

If you consider ax + b mod c to be "extremely complex."

Making (and testing) a random number generator is a standard assignment in programming 101 classes.

This reminds me of something I did in my first year in a computer science course. I used the C rand() function on a Unix server to generate triplets - an X-coordinate, a Y-cordinate and a printable ASCII character, which it then plotted in a window. I then told it to continue running until I stopped it. I used (rand() * size) and truncated the result (rand() generates a floating point number between 0 and 1). Interestingly, it produced patterns on the screen. The random printable character seemed random, but the X/Y pairs tended to form a grid pattern - it was definitely NOT random.

Since then, I've had a healthy disrespect for random number generators. For most uses, they're fine, but there's always the danger that they'll have patterns to the numbers they generate (and shown by the pattern the UNIX rand() function was producing).

I did write a variant that appeared to be random, but it wasn't out-of-the-box.

Yeah, I just re-read your post in more detail and realised that, too :)

Not having the required CL just means +5 DC. A monk can still craft them - it's just a bit more of a challenge. You still need the appropriate feat, though, and the appropriate skill (depending on what you're making). All other missing prerequisites just add five to the DC.

EDIT TO ADD:

See this comment from Sean K Reynolds on the subject

Anything that bumps Charisma or Charisma checks.

eg. Circlet of Persuasion

Assuming this hair is the only natural weapon on your witch, it will be counted as a primary natural attack that deals 1d3 + 1.5x STR modifier in damage (assuming your witch is medium size). If you have claws or some other form of attack, there is no way for it to become primary -- that is, of course, you don't want to amputate your anatomy.

Assuming you have more than one instance of the 'hair' natural weapon type/form, you can have each hair natural attack count as primary attacks that deal 1d3 + STR modifier. How you could possibly get more than one instance of a natural hair attack, I don't know.

This is as per the PRD rulings.

The Multiattack feat allows you to reduce the penalty on secondary attack, but the pre-requisite pretty much requires you to have three or more natural attacks.

That feat is generally used for creatures like horses that have primary and secondary natural attacks, but the feat itself doesn't require it. As far as I can see, if you have the feat, you could use weapon attacks and natural weapon attacks, with the natural weapons taking a -2 penalty instead of -5.

WRoy, I understand what you're saying, but let's look at AoO again. You can get an AoO if a foe leaves a square that you threaten

Threatened Squares: You threaten all squares into which you can make a melee attack, even when it is not your turn. Generally, that means everything in all squares adjacent to your space (including diagonally). An enemy that takes certain actions while in a threatened square provokes an attack of opportunity from you. If you're unarmed, you don't normally threaten any squares and thus can't make attacks of opportunity.

If you have armor spikes/spiked gauntlet with a reach weapon, you can make a melee attack at 5' because of the FAQ I linked to. That means you threaten that square, and that means you can make an Attack of Opportunity.

A little convoluted, but it appears to be RAW to me. The free action required to release the hand on the two-handed weapon is enough to allow you to threaten that square, because the threat is determined by what were you are able to make a melee attack, not by whether or not it is your turn.

Berdache wrote:

We play the 1 = fail on skill checks but I have never really liked it. Image a skill check so trivial that anyone (even unskilled) would normally succeed doing it on a 1. So in combat a party of 4 adventurers will on average fail this check every 5 rounds? This leads to some silly situations.

A bit off topic, but I have to fix your math here.

Actually if it was 1 in every 5 rounds that's a 20% chance of failure.

A mess up of 5% or a 1/20 is 1 time in every 20 rounds. Much more realistic.

Think of even professional athletes who are super specialized in what they do. No one is perfect EVERY time, there is always a chance they screw up, hence the 5% chance they mess up on that "1."

The same can be said for that little kid throwing the basketball at mid-court. He would miss that shot almost every time, but that one time he does it's awesome hence that 20 succeeds and that 5% comes to fruition.

These are the reasons that I house-rule those skills in my game, but I understand that others don't think the same way as I do.

Yeah, but those are more 1% (or lower) chances, not 5%. 5% is too high for that sort of event.

But we're not talking about two weapon fighting - we're talking about whether you have the choice of which weapon to use to attack.

Do you have the choice to use the reach weapon or the armor spikes/spiked guantlet at the time of the potential AoO, or do you have to make the choice at the end of your turn (and hence that choice would be applicable for any AoO opportunities that could arise)?

As far as I can tell, there's nothing in the rules that gives a clear indication either way. This FAQ implies that you can make the choice at the time of the AoO opportunity, since it's a free action to free a hand so you can use armor spikes/spiked guantlet. That's the way I'd rule it, but that's just my interpretation.