Sorry perhaps I was too snarky. I’ve heard too much “dElVer isNT a pRoBlem, iTS jUsT MtGo mEtA” over the years.
I am using the statistical definition of significant: not due to chance. The actual observed difference is in the numbers on the left.
People want to feel like they can beat the best deck if they play tighter than the opponent. The stats bear out that it isn’t true (-EV is about the same as the + EV of being a good player). But there is also the intangibles you mention. Magic is so full of haymakers nowadays and games end immediately. When the delver opponent plays an unkillable 8/8 on t3 and always has the game ending hate card because of EI + DRC you can’t kid yourself that maybe you could have played it better. You lost because they are playing the WOTC $$$$ Mythic and you aren’t.
They forced your answer to murktide? Nothing you could have done. You’re dead.
They always find their 2-of meltdown by t3 against your shortcake deck? Nothing you could have done. You’re dead.
That’s the sort of intangibles that makes people quit or play delver.
This is a big problem for me too. The fact that Delver can super-efficiently parse through half of their deck per game means that it feels like you get slammed with really unfair shit like Meltdown, Ruination, Price of Progress, etc far too often. I also come from a time when blue would never get something as efficient as Murktide Regent. To me Delver is much worse when they have to splash for stuff like Goyf. It's all been said before, it is what it is for now. The timing feels right schedule-wise for a B&R announcement, fingers crossed.
Like I just almost won the challenge because no-one expected a deck to try to go under delver and Humans was mostly immune to end the festivities. But in semis Reid Duke quickly dug through half his deck to find his 1-of rough//tumble, and I lost. I’m sure that once players see my great finish where I didn’t drop a game against fair blue all swiss, they add 2 rough//tumble and I have a losing record. Because they will always have it when they can see 1/3 of their deck by t3 while committing threats, answering yours, and accruing card advantage at the same time.
Added bare bones doomsday and GW Depths primers.
There's a reason all the good cantrips are banned in Modern.
Which is why cards like Murktide and Delver are idiotic. Why does Blue have the best beaters?I also come from a time when blue would never get something as efficient as Murktide Regent. To me Delver is much worse when they have to splash for stuff like Goyf. It's all been said before, it is what it is for now. The timing feels right schedule-wise for a B&R announcement, fingers crossed.
If there's any bans I would prefer it would be Murktide and Delver. They can always do it one at a time if they want to be more conservative about it.
I don't think its one or the other. It's a problem when Blue can see 3-4x the amount of cards per turn. It's a problem that fetches exacerbate that and make Delve more attractive. And its a problem that Blue has the best aggressive creatures.
They obviously will never get rid of fetches or cantrips from Legacy so then you ask what might they actually do?
I'll try to avoid further muddying up this thread with Ban discussions since there's a separate thread for that...
Did people find my post useful? Should I post an updated one?
Anyone know a good bit about HTML or something? Would it be possible to export the structure of one of the MTGGolfish entries to here (I can rehost the card images easily) in a somewhat easy manner? This would be an easy way to make threads for "new" decks.
"The Ancients teach us that if we can but last, we shall prevail."
—Kaysa, Elder Druid of the Juniper Order
Wait, are you asking to recreate a webpage Ina post?
My interest was piqued so I took a look at this site a little bit, as a web dev myself I feel pretty good about saying that this content is not easy to get at if you're not familiar with html.
I didn't go suuuper deep but I checked the Network tab and the Elements, I have a strong suspicion that the decklist and its display elements are server-side generated (so there's not like a "decklist.json" sent to the client that is interpreted and rendered, which makes sense for a site like this but is inconvenient). You'd go a little crazy grabbing the elements yourself, you'd probably want a scraper to grab the parent element and just yeet the desired text of its children into a file.
This is an assessment after like the briefest of glimpses into how that site works so if one is you is galaxy-brained and sees something I didn't, forget I was even here
More like use the structure to format a post.
What I am really after is less the decklist and more the "X copies in XX% of decks" part. Perhaps it would just be easier (since I really know nothing of HTML) to copy it manually. I'll have to take a more detailed look when I have a little more time. I look at the source too and, well, since I don't know anything, it was not particularly clear to me.
"The Ancients teach us that if we can but last, we shall prevail."
—Kaysa, Elder Druid of the Juniper Order
Ah ok, that output is actually a lot more sane than the decklist page for whatever reason. Just looks like a couple of tables with text more or less as you'd expect (apart from the mana symbols, those are svgs)
Depending on what it is you're actually looking to do, it might just be a time-saver to c/p what you see on the screen. If I do that I get a pretty straightforward output (minus the analysis symbols but it sounds like you don't need that for what you're doing)
So I still don't know what you're asking for. If you're painstakingly clear (ie step by step what you want to do, never using words like or that and always saying the whole name of the noun you want to work on) I can probably do it for you.
Anyways scaping off goldfish isn't that hard: First we right click and select inspect. On chrome it actually highlights the rendered element (left) when we mouse over the code:
Here you can see I have my mouse over the <table class> element, specifically named "deck-view-deck-tale".
Clicking the down arrow will expand the list and you can see inside the <table class> is a whole bunch of <tr> (tale row) elements. Some of these are headers, you can see those are marked because have names provided: <tr class "deck-category-header"> We don't really care about those. If you keep drilling in an unlabeled <tr> you'll see it contains four table data <td> elements. The first two are the parts I think you want: The quantity and the card's name.
So here's the entry for the 1 spell pierce:Code:<tr> <td class="text-right"> 1 </td> <td> <span class="card_id card_name"><a data-card-id="Spell Pierce [NEO]" data-full-image="https://cdn1.mtggoldfish.com/images/gf/Spell%2BPierce%2B%255BNEO%255D.jpg" rel="popover" href="/price/Kamigawa+Neon+Dynasty/Spell+Pierce#paper" data-original-title="" title="">Spell Pierce</a></span> </td> <td> <span class="manacost" role="img" aria-label="mana cost: blue"><!--?xml version="1.0" encoding="utf-8"?--> <svg viewBox="-420 0 100 100" xmlns="http://www.w3.org/2000/svg"> <g> <circle fill="#C1D7E9" cx="-370" cy="50" r="50"></circle> <path fill="#0D0F0F" d="M-352.512,83.719c-4.787,4.871-10.684,7.307-17.688,7.307c-7.861,0-14.098-2.69-18.711-8.073 c-4.359-5.127-6.537-11.662-6.537-19.606c0-8.543,3.717-18.286,11.15-29.224c6.064-8.969,13.199-16.83,21.402-23.58 c-1.197,5.469-1.793,9.355-1.793,11.662c0,5.299,1.664,10.467,4.996,15.508c4.102,5.98,7.219,10.426,9.357,13.328 c3.332,5.043,4.998,9.955,4.998,14.737C-345.336,72.871-347.729,78.852-352.512,83.719z M-352.641,56.357 c-1.281-2.861-2.777-4.762-4.486-5.703c0.256,0.514,0.385,1.24,0.385,2.18c0,1.795-0.512,4.357-1.539,7.689l-1.664,5.127 c0,2.99,1.492,4.486,4.484,4.486c3.16,0,4.742-2.095,4.742-6.281C-350.719,61.721-351.359,59.223-352.641,56.357z"></path> </g> </svg></span> </td> <td class="text-right"> $ 0.21 </td> </tr>
One table row (the text between <tr> and </tr>)
One table data for the quantity:
(there's this thing called CSS where if you name an element a style your browser will "just know" how to render it (as right aligned text, in this case)Code:<td class="text-right"> 1 </td>
One table data to wrap around a class to define the card's info:
One table data to define the cost:Code:<td> <span class="card_id card_name"><a data-card-id="Spell Pierce [NEO]" data-full-image="https://cdn1.mtggoldfish.com/images/gf/Spell%2BPierce%2B%255BNEO%255D.jpg" rel="popover" href="/price/Kamigawa+Neon+Dynasty/Spell+Pierce#paper" data-original-title="" title="">Spell Pierce</a></span> </td>
Now there's some more bullshit there, but you don't care how it rendered or found the image, so I snipped it out for now.Code:<td> <span class="manacost" role="img" aria-label="mana cost: blue">_</span> </td>
And finally the price:
And that's how they format their deck data. Tsumi is correct that all this data is gathered server side, but it all has to be sent to you in order to render.Code:<td class="text-right"> $ 0.21 </td>
Yeah I was more hoping for some JSON blob that could be parsed by a script and formstted easily instead of reading some automatically-generated HTML, but fortunately just copying and pasting the actual text on-screen without using the Dev Tools gives a friendly enough output that unless you're looking to do this like 10 times a day I don't think there's a reason to scrape or inspect anything.
It's this part that I am after:
The Decklist section isn't what I'd need to really copy from there, I can find my own decklists to paste in a post. What I want are the card categories (Creature, Planeswalker, etc.) and the "X copies in XX% decks." The reason I had the idea of trying to take the HTML is so that the post would have that sort of format, with the text centered under each card image. Rehosting the images is easy enough for me, I have a browser extension that can do it in one click.
But to get the text under each image, while keeping it centered with each, is less immediately easy for me. I probably just need to dig a little deeper on the "Inspect" page for each of those elements though.
"The Ancients teach us that if we can but last, we shall prevail."
—Kaysa, Elder Druid of the Juniper Order
Ah, well this isn't really an issue of HTML, what you want is something like react or javascript that when the page is loaded will do these maths for you and then output some HTML to make it look the way you want.
Above is the element of those cards in Izzet murktide, as you can see it doesn't use any calculations it just looks to output "4.0 in 100% of decks".Code:<div class="spoiler-card"> <a href="/price/The+List/Dragons+Rage+Channeler"><picture> <span class="price-card-invisible-label">Dragon's Rage Channeler</span> <source srcset="https://cdn1.mtggoldfish.com/images/gf/Dragon%2527s%2BRage%2BChanneler%2B%255BPLIST%255D.jpg" type="image/jpg"> <img class="price-card-image-image" alt="Dragon's Rage Channeler [PLIST]" src="https://cdn1.mtggoldfish.com/images/gf/Dragon%2527s%2BRage%2BChanneler%2B%255BPLIST%255D.jpg"> </picture> </a> <p class="archetype-breakdown-featured-card-text"> 4.0 in 100% of decks </p> </div>
When the page is loaded, the server will do those calculations and tell your browser "Hey, write this down: 4.0 in 100% of decks"
The Best Practices way to achieve your goal would be to have an actual database of all decks, so that you could select/aggrigate decks by a criteria and then filter those results.
Something like:
Deck ID, Deck Archetype, CardName, Quantity, Date
And then use SQL to querry this database of all decks
select CardName avg(Quantity) from DeckaBase where Archetype=Delver group by CardName
which will output somehting like
Delver 4
DRC 4
ect ect
and then you take those outputs and use them as inputs for whatever you're using to make your webpage
There are currently 1 users browsing this thread. (0 members and 1 guests)