18PA first play test

[Walt Collins]

We didn't get to finish last night, but we got as far as the formation of the NYC.  We will likely be playing again tonight, so I will offer only my preliminary observations here, which may be confirmed or disabused with further plays.

1. We had some track shortages, and needed to add a couple of straight yellow cities (#57), 3-way green "Mercedes" (#81), and green left turn and right turns (#82 and #83).

2. The facts that there are no non-straight yellow cities, nor any tight curved yellow track, nor any brown track upgrades, seem to make it very possible for companies who start after the first OR to be severely damaged in their run capabilities.  I will report back on this after more testing.

3. The Local Companies are not close to equal, but they all have the same price, and are allocated by stock turn priority seat only.  By pure luck I wound up with #5 (LIRR) and #6 (OCRR) and they were almost immediately paying out $45/round.  Company #7 (B&M) has a problem if it doesn't build quickly as soon as the green tiles are available, and can easily be blocked.  If it doesn't get blocked, #7 will run along the ferry and down to Islip for $60/round.  Companies #8 (NYHRR) and #9 (PATH) seem by far the weakest in revenues, although presumably those tokens would be worthwhile in the brown and gray phases.  However, #5 also has a NY token and still gets excellent revenues.  Perhaps these Local Companies could be included in the initial auction, but not be required to be purchased?

4. The 2x tokens are very powerful.  The 4 LGCs automatically get one for free when they destinate.  And there are 2 more in Scranton for anyone to grab.  Erie has a huge advantage in being only 3 tiles away, so it can grab a Scranton token on its first OR.  That leaves Reading and PRR to fight for the other one (if Reading is even opened early).  Doubled city value, plus an additional +$10 bonus, makes Scranton worth $70 in the green.  Scranton sits in the center of the board, and it has 4 exits in the green, so it can easily be used by both a company's trains.  Since it's right on the way to Buffalo, it makes Erie super powerful, as it can run a 2T from Scranton to Newark for $100, plus a 3T from Scranton to Buffalo via Rochester and 3 towns for an additional $210.  No other company on the board can make that much in the early mid-game.  Plus it can use both it's 2x tokens with multiple trains because they're not off-board.  The PRR and B&O destination tokens are lower value ($30 instead of $40 for Buffalo) and can only be used once each per round.

5. The B&O major seems severely handicapped.  It's not likely to be in the running for a Scranton token.  The best it can hope for is laying a token in Philly, and that's saying something. ;)

The B&A didn't come out until late in our game, and never developed any track.  The NYC only operated once, and its token locations seemed lackluster.  We'll see how those develop in future sessions.

Walt

[David Hecht]

Well, I appreciate your report, though I have to say that it's generally the case in my games that the endgame counts for a lot. You describe the NYC as lackluster...there was a longish period in which the NYC was the company whose control was most correlated with winning: YMMV.

The track shortages you describe--are they based on the printed tile mix? Because if so then you need to take cognizance of the revisions. In particular, there are seven (not three) #57 (straight yellow city) tiles, and there are four #81 (Mercedes), ten #82 (right turn), and six #83 (left turn) tiles.

You may wonder why the left and right turns are asymmetrically distributed: in the games for which I've had data collected (either by me or by others), there's always a far greater demand for #82 than #83. It's a mystery to me too.

I don't doubt it's possible for people to get screwed around by tile lays. So far, no one else has had any issue with this. Yes, it's nice to make the board totally interconnected: I don't like it. I think players should be required to think a bit about their tile placements and upgrades. I'll note for the record that 1830 had only straight yellow cities and no one ever complains about that.

The local companies are by definition not all equal. However their inequality is time-phased. You undervalue #8 and #9 because you didn't play out the game. Furthermore, once the local companies turn into 2R-trains, they are in fact equal. The weakness of #5 in this regard is that it's very hard to link to, once phase 5 drops and the Fall River Ferry falls off.

With the x2 tokens, again...need to play out the game. I do think I should tweak up the PIT and COL off-map locations: possibly 20/40/60 instead of 10/30/50. But that's a smaller issue. Yes, x2 tokens work well; yes, the Erie can get to Scranton before anyone else (if it moves first, several companies can beat it there using the #0 special power). However you should also ask yourself if it makes sense to put an extra $40 in the hands of the #4 owner, assuming it isn't the same guy (which obviously you can preclude, since it's very unlikely that the #4 owner will also have priority in the ISR). When the ownership is split and the Erie ignores Scranton, sure--it runs a bit less well. But the owner of the #4 is disadvantaged too. And of course if you load up on 2-trains, you're likely to be going backwards when the 4-trains drop.

Finally as to the B&A, if you start it out of the gate, it can destinate in two ORs even if it chooses to buy a train (and leave building status) in OR-2. Then it can run two 2-trains out of Hartford to Albany and Boston, which I think you'll find is quite respectable.

All these companies have strengths and weaknesses. The mark of a successful design (if I may be so self serving) is that it's not apparent what those are at first. Referring back once again to 1830, the PRR has perhaps the worst initial start (with the obvious exception of the Erie), yet the stats show that something like 40 percent of the winners own the PRR. Not exactly obvious at first bite, I can tell you.

Again, don't mean to rain on your parade or to suggest you shouldn't bring these things up. But I'll take your suggested design changes a lot more seriously when you've played a dozen or so games.

--
You received this message because you are subscribed to the Google Groups "dtg-proto" group.
To view this discussion on the web visit https://groups.google.com/d/msg/dtg-proto/-/z4uwdE0WzkAJ.
To post to this group, send email to dtg-...@googlegroups.com.
To unsubscribe from this group, send email to dtg-proto+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/dtg-proto?hl=en.

[Walt Collins]

I saw that you had asked somewhere about how many tiles we found we needed, so I just put that in as a mention, hoping to confirm the findings you'd seen from other games.  This was prior to adjusting the tile mix, which I've done today as per your specifications, so I'm sure the new tile mix will be better, as the specific shortages I mentioned were already addressed.

I look forward to giving the B&A an early try, as well as seeing how the B&O fares in the later game, and how #5-#9 Local Companies influence phases 5 and 3D.  Hopefully we'll get to those this evening.

Walt

[David Hecht]

Thanx. BTW, when I ask about tiles, I mean an actual count of what's on the board at the end of the game. General statements are helpful but need to be refined: "a couple of" can mean two, or it can mean more than that. Which of those it is is important! :-)

I wanted to expand a bit on the situation in the southwest. There's a lot of companies competing for relatively limited real estate (just like in real life!) and I've gone back and forth on both the CNJ and the RDG on whether they should be full-on companies or locals. One thought I've had (anyone can chime in here...Tom McCorry? You listening to this? <vbg>) is to reduce the RDG to a local, rebase it in PHI, and reduce Reading back to a town. As part of this, I'd do a pre-printed yellow PHI (worth $30) that has three one-ended stations pointing E, NW (RDG would be here), and SW: similar to the layout of the EWR tile. Green PHI would be worth $40 and add a second exit across the tile for each station (exactly like tile #190, the ATL tile from 1832). Brown and grey PHI would be as now, $60/$70 with three concatenated city circles. This would have the overall effect of decongesting a bit, and help out the CNJ and B&O (who would be virtually guaranteed spots in PHI, whereas the PRR would only be guaranteed one if it ate the Reading).

The main reason I don't do it is, what would replace the Reading as the eighth company? I really dislike games with odd numbers of companies (and, yeah, that includes my 18VA...), but I've about exhausted the plausible candidates. Plus even if I added one, where would it go? You see the problem.

But if you (or any other participant...) has any thoughts on the subject...

To view this discussion on the web visit https://groups.google.com/d/msg/dtg-proto/-/uSDIq2Jd6fUJ.

[Chris Shaffer]

Why do you dislike games with odd numbers of companies?

--
Chris

Please consider the environment before printing this e-mail.

[David Hecht]

The obvious one: with few exceptions, in 18xx games the player with only one company is severely disadvantaged.

Hm. Maybe I'll dust off the Transit Authority...

[John A. Tamplin]

On Thu, Oct 18, 2012 at 3:49 PM, David Hecht <bar...@earthlink.net> wrote:

The obvious one: with few exceptions, in 18xx games the player with only one company is severely disadvantaged.

Hm. Maybe I'll dust off the Transit Authority...

But an odd number doesn't have much to do with that -- if you have 9 companies and 3 players, then everyone could have the same number.  I can't think of any games where the number of major companies is divisible by every number of players, so someone is always at risk of getting screwed.  If you think that matters, then you make arrangements to get one (assuming the game has ways of manipulating priority besides always sitting downstream of Mark :).

--
John A. Tamplin

[Mark Geary]

On Thu, 18 Oct 2012, John A. Tamplin wrote:

> But an odd number doesn't have much to do with that -- if you have 9 companies and 3 players, then everyone could have the same number. I
> can't think of any games where the number of major companies is divisible by every number of players, so someone is always at risk of getting

Obviously, someone needs to design a game with 60 companies.

Mark G

--
Don't forget to stop and eat the roses.

[David Hecht]

The issue isn't about different numbers of companies per player as such: it's about players having one versus two companies. Having only one company in a game where other players have two is problematic. In fact, you were a participant in a playtest of 18Ardennes where we had to address the ultimate version of this problem: in a five-player game, the player who got the sixth (and last) company had a huge advantage. We instituted the auction system for just that reason.

And of course, the more companies total, the less it matters. You might argue that a game with 23 companies won't play very differently than a company with 24: and you'd be right. But I think there's a tipping point below which it matters greatly, and I think eight is it. Just think about the huge differences in games with eight versus six major companies: the former are treated as "normal" whereas the latter almost always have to have a whole host of special rules to balance them.

[John A. Tamplin]

On Thu, Oct 18, 2012 at 4:37 PM, David Hecht <bar...@earthlink.net> wrote:

The issue isn't about different numbers of companies per player as such: it's about players having one versus two companies. Having only one company in a game where other players have two is problematic. In fact, you were a participant in a playtest of 18Ardennes where we had to address the ultimate version of this problem: in a five-player game, the player who got the sixth (and last) company had a huge advantage. We instituted the auction system for just that reason.

And of course, the more companies total, the less it matters. You might argue that a game with 23 companies won't play very differently than a company with 24: and you'd be right. But I think there's a tipping point below which it matters greatly, and I think eight is it. Just think about the huge differences in games with eight versus six major companies: the former are treated as "normal" whereas the latter almost always have to have a whole host of special rules to balance them. 

Ok, but the point is that "odd" isn't what you care about, it is having at least 2n companies. 

--
John A. Tamplin

[David Hecht]

On 10/18/2012 4:46 PM, John A. Tamplin wrote:
> Ok, but the point is that "odd" isn't what you care about, it is
> having at least 2n companies.
>

Yes. I assumed that in the present context, there would be no need for
that to be unpacked.

My mistake.

[John David Galt]

> On Thu, 18 Oct 2012, John A. Tamplin wrote:
>> But an odd number doesn't have much to do with that -- if you have 9 companies and 3 players, then everyone could have the same number. I
>> can't think of any games where the number of major companies is divisible by every number of players, so someone is always at risk of getting

On 2012-10-18 13:35, Mark Geary wrote:
> Obviously, someone needs to design a game with 60 companies.

1841 v1 comes close to that ideal (though I've seen it run out, even so).

[John David Galt]

On 2012-10-18 13:37, David Hecht wrote:
> And of course, the more companies total, the less it matters. You might argue that a game with 23 companies won't play very differently than a company with 24: and you'd be right. But I think there's a tipping point below which it matters greatly, and I think eight is it. Just think about the huge differences in games with eight versus six major companies: the former are treated as "normal" whereas the latter almost always have to have a whole host of special rules to balance them.

When I see a game with only 6 major companies I assume that at most 3 should play
(or at least, were intended to play).

[John David Galt]

On 2012-10-18 12:49, David Hecht wrote:
> The obvious one: with few exceptions, in 18xx games the player with only one company is severely disadvantaged.

This is why I felt the need to add a #16 to 18EU (and a couple of extra major
companies don't hurt either, when playing with 5 or 6 people).