I’m preparing for a test and I keep hesitating when I see a quadratic because I’m not sure which method to use. I can factor, complete the square, and use the quadratic formula, but in practice I second-guess the choice. For example, with an equation like 6x^2 − 5x − 4 = 0, I can’t tell quickly if it’s factorable, and completing the square seems messy with fractions, so I default to the formula and then worry about sign mistakes.
What’s a practical, step-by-step way to decide which method to use under time pressure? As a follow-up, if the coefficient of x^2 isn’t 1, is it better to complete the square directly or try to simplify first (like factoring out a common factor)? Also, is there a quick check for factorability-like using the discriminant-that’s actually worth doing during a test?
Ready to make maths more enjoyable, accessible, and fun? Join a friendly community where you can explore puzzles, ask questions, track your progress, and learn at your own pace.
By becoming a member, you unlock:
3 Responses
Great question! Under time pressure, I use a quick decision roadmap: first scan for a common factor and any obvious pattern (difference of squares, perfect square), then do a 5-second discriminant check D = b^2 − 4ac. If D < 0, stop (no real roots). If D isn’t a perfect square, don’t waste time factoring-go straight to the quadratic formula. If D is a perfect square, it will factor over the integers, so try the ac-method (find integers with product ac and sum b) or a quick rational-root try. Completing the square is best when a = 1 and b is even (or after dividing by a to make it monic); otherwise it spawns fractions, so prefer the formula. Always factor out a greatest common factor first to simplify everything; if a ≠ 1 and there’s no GCF, I rarely complete the square unless I need a vertex form. To reduce sign slips with the formula, compute D first, then plug into x = (−b ± √D)/(2a) as a single template. Example: 6x^2 − 5x − 4 = 0 has D = 25 + 96 = 121 = 11^2, so it should factor; since ac = −24 and we need sum −5, use −8 and 3: 6x^2 + 3x − 8x − 4 = 0 ⇒ (2x + 1)(3x − 4) = 0, giving x = 4/3 or x = −1/2. That same D also makes the quadratic formula quick and confirms the result, so you’ve got a tidy cross-check!
My test-day triage goes like this: first, clean the problem-factor out any common GCD so the numbers are smaller. Next, spend (truly) 10 seconds looking for factors: if it’s monic (a=1), hunt for two integers that multiply to c and add to b; if not monic, try the “ac” trick-look for two numbers with product a·c and sum b, then factor by grouping. If nothing pops quickly, compute the discriminant D = b² − 4ac: if D is not a perfect square, stop factoring and use the quadratic formula; if D is a perfect square, it will factor over the integers, so you can either factor or use the formula and get tidy rationals. Completing the square is great when you need vertex form or when a=1 and b is even; if a ≠ 1, I’d factor a out of the x-terms first (write ax²+bx as a[x² + (b/a)x]) and complete the square inside-this stays clean only if b is a multiple of 2a; otherwise it’s fraction city and I’d skip it under time pressure. As a quick sign-error guard with the formula, compute D separately, then do x = (−b ± √D)/(2a) with the numerator evaluated before any simplifying-I even whisper “minus b” to myself so I actually flip the sign. On your example 6x² − 5x − 4 = 0: ac = −24, and numbers −8 and 3 fit (sum −5), so 6x² − 5x − 4 = (3x − 4)(2x + 1), giving x = 4/3 or −1/2; equivalently D = 25 + 96 = 121, so x = (5 ± 11)/12, same roots. So, short version: GCD clean → quick factor sniff → discriminant decides factorability → formula if D isn’t a square (or if time is tight), and complete the square mainly when the setup is friendly or the question asks for it. I’m slightly hand-wavy on the 10-second cutoff, but that rhythm has saved me a lot of eraser crumbs.
Here’s a fast, no-drama playbook I use under time pressure:
– First, yank out any common factor. If that makes a=1, great-it’s factoring time. If not, spend 5–10 seconds tops on the ac trick: compute ac, scan factor pairs. If ac<0, look for a pair with difference |b|; if ac>0, look for a pair with sum |b|. If nothing pops, stop. Next, check the discriminant D=b^2−4ac. If D<0, move on (no real roots). If D is a perfect square, it will factor over the integers, so do it. If it isn’t, go straight to the quadratic formula-don’t burn minutes forcing factors. For the example 6x^2−5x−4: ac=−24, pairs 8 and 3 differ by 5, so factor: 6x^2−5x−4 = 3x(2x+1)−4(2x+1) = (3x−4)(2x+1). Also D=25−4·6·(−4)=25+96=121=11^2, so the perfect-square check agrees. Easy. For the formula, compute D first, then do x = (−b ± √D)/(2a). Writing it that way cuts sign errors in half.
About completing the square: it’s great for vertex form or when a=1 and b is even (clean halves). If a≠1, I usually don’t complete the square unless dividing everything by a keeps the numbers friendly; otherwise the formula is faster. You can’t really complete the square cleanly when a≠1 without making a mess, so I’d only do it after simplifying first. One more quick tip: if D=0 you get one repeated root, so there’s nothing to factor into two distinct binomials-just use the formula and move on. For more examples of choosing a method, this rundown from Khan Academy is solid: https://www.khanacademy.org/math/algebra/x2f8bb11595b61c86:quadratics
Curious: when you hesitate, is it usually because ac has many factor pairs, or because the numbers are big and you’re worried about arithmetic slips in the formula? If it’s the latter, we can set up a couple of “muscle-memory” checks to strip out sign mistakes.