Time for a break from serious posts & deep thoughts. It's game night!
I am not the most proficient Magic: The Gathering player. I do not have an encylopedic knowledge of cards, combo pieces, and keywords. I do not have the best ability to glance at a card and see whether it's weird rules are jank or a powerhouse piece. I am a thoroughly casual player, and totally OK with that status.
However, I do like to win at least every now and again. And that requires either opponents who play on autopilot, or having a reasonably coherent deck to deal with their shenanigans. Here is my process for kludging a deck together and refining it. Is it the best process? No. But I don't want to "net deck," and I don't have the budget for pay-to-win craziness. This is just what works for me.
No matter the format (Commander, Pauper, Modern, etc.) I usually start by looking through my collection for cards or a mechanic I want to use as the foundation.
Then, I start with a land base of about 40%, so 38-40 lands for Commander and 24 for any 60-card format. Build the mana base first, and lock that in. The easiest way to ruin a functional casual deck is to look at the mana base for cEDH or competitive Modern, and assume that'll work for you. If you find yourself "mana flooded" most of the time after testing for several games, then consider cutting lands. Not before.
Finally, I discover half the cards I threw together in the deck at first don't actually work toward the same goal, which forces me to choose one path to rebuild toward one cohesive game plan. Like I said, I don't have encyclopedic knowledge of the game, and quickly veer from "have an alternate win condition" into "incoherent mess."
I'll use my new K'rrik, Son of Yawgmoth deck as the perfect example of tuning a pile of cards into a reasonably capable deck. There is still room for improvement, but this is officially "good enough." I wrote about it twice before, but I have made a lot of refinements. This is how I fixed that mess, and the tools I use along the way.
The official database of Magic cards is called Gatherer, and it has several useful tools to find cards in its advanced search. That site is also the source for the image of the K'rrik card, by the way.
I needed to find cards which might add benefits for me when I pay life to cast cards with K'rrik's ability, so I chose the advanced search, typed "lose life" in the rules search box, and chose to narrow results to cards using only black mana. There are over 30,000 cards now, but this narrowed it down to just 43 cards.
Not all were useful, and of those which had promise, I didn't always have a copy, but I knew I had a copy of Vilis, Broker of Blood in my collection. It was still in my Queza deck, and I swapped it for Syr Konrad the Grim since each seemed better suited to the other deck.
With so many cards published over the years, Magic has many cards with similar or even duplicate effects. Sometimes the same effect was printed on different cards for a different mana costs, or extra bonus effects, several years apart. If I have a card I like, but want more ways to achiefe its effect in a singleton game like Commander, or want an upgrade to a playset for a 60-card deck, Strictly Better can be incredibly useful.
Sign in Blood (picture again via Gatherer) is a classic card draw spell in mono-black. It has been in print for over a decade. Painful Lesson has exactly the same effect, but it costs one black mana plus two generic mana of any color. This is strictly worse... unless my deck has ways to make colorless mana, or I am playing more than one color. I had forgotten this card even existed until I searched for Sign in Blood on Strictly Better to make sure I wasn't missing anything.
Some discretion is necessary, though The site says Deadly Dispute is strictly better, probably because it only requires one black mana and nets a treasure token as a bonus, but I would need to sacrifice a creature instead of paying life. This deck is not built around sacrifice mechanics, so that advice is disregard. In fact, this deck had a lot of such "aristocrat" effects I purged in my effort to focus on a single main theme. If I get around to building one, though, Deadly Dispute will definitely belong there.
One of the most popular (and most misused IMHO) tools is a site called EDHrec, which offers recommendations for Elder Dragon Highlander (EDH), the old community name for Commander. This site trawls all the sites where people share their decks and aggregates what is popular. If I plug in my deck list, it will suggest what to add, what to cut, and what I would expect to pay for those other cards.
However, this comes with a big caveat: popularity does not prove a card is good. The EDHrec feedbck loop can result in poor synergy if popularity of a mismatched card for your deck has led to too many online lists including it. Unpopular cards can also be unsung heroes in the right deck and budget level. There is also a bias toward net decks which no one actually plays, but have been built as digital experiments.
Can you build a deck by plugging in a commander, picking the top 60 cards it recommends, adding some lands, and playing a game? sure, but it will be expensive, and not really guaranteed to work. You'll also pay a lot for expensive staples that may not really help you win.
It's not all bad, of course. This is truly a useful site, and we can just use Sign in Blood again to demonstrate another neat feature. That link shows combos and similar cards I can consider using to make my deck more synergistic. This is also good for any format where the card is legal, not just Commander/EDH.
After some relatively recent updates, this site went from leaning more toward a joke to an actually useful tool for analyzing the power of my decks. The original focus was on how angry your deck would make your opponents, or how salty it was, if you don't get the slang. I appreciate the creator's disclaimer/F.A.Q.
Is the site 100% accurate and infallible? Emphatic NO. Am I a world-leading expert in cEDH design philosophy? Again, emphatic NO.
I'm just a guy that loves to code and loves the game. It is sincerely funny to me how much hate the site gets in Reddit and Discord — it's one thing to celebrate salty deck building, but to literally make people salty just by existing? Amazing!
However, after a revamp, this is actually packed with useful tools. He has a page covering his algorithm and methodology, so you can see how he reaches his conclusions.
I hope you can see its analysis for my deck. It can list combos, expected average mana production, and warn if the deck is too unfocused withd ecent accuracy. Is it the best such tool? I have no idea. Does it have an avatar of a neckbeard who roasts you in the bottom left corner of the screen? Absolutely. And it made that handy rule-zero card for the image here, too.
That site says my deck is roughly bracket 3, and a 6.9 on the old 1-10 power level curve no one ever really defined. I even tried, poorly. There are other sites which try to analyze deck power and bracket, of course. This one says mine is a 5.4, bracket 4 deck, thanks to the speed at which it could win via combo. This is why Rule 0 matters. Have an honest discussion with other players about your deck, and be lenient when something doesn't slot neatly into a given definition.
Enough talk. Time to go play some games. The latest changes I made need testing in the real world, not just on computer spreadsheets! Are you ready for this, 
If you're not on Hive yet, I invite you to join through InLeo or PeakD. If you use either of my referral links, I'll even try to delegate some Hive Power to help you get started.
